Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Multiple roles for integrins in Drosophila glial development Xie, Xiaojun 2012

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2012_fall_xie_xiaojun.pdf [ 6.35MB ]
JSON: 24-1.0072946.json
JSON-LD: 24-1.0072946-ld.json
RDF/XML (Pretty): 24-1.0072946-rdf.xml
RDF/JSON: 24-1.0072946-rdf.json
Turtle: 24-1.0072946-turtle.txt
N-Triples: 24-1.0072946-rdf-ntriples.txt
Original Record: 24-1.0072946-source.json
Full Text

Full Text

MULTIPLE ROLES FOR INTEGRINS IN DROSOPHILA GLIAL DEVELOPMENT  by Xiaojun Xie M.Sc., University of Western Ontario, 2007  A THESIS SUBMITTED IN PARTIAL FULLFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in The Faculty of Graduate Studies (Zoology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) July 2012  © Xiaojun Xie, 2012  Abstract Glia are well known for providing essential physical and metabolic support to neurons, as well as regulating neuronal development.  Glial development is also  modulated by external signals from other cells and the extracellular matrix (ECM). Many signals are transduced into glia by specific receptors, such as integrins for the ECM. Previous studies show that integrins are expressed by all major vertebrate glial subtypes and play key roles in many important developmental processes.  However complex  composition of the integrin family and difficulties of manipulating genes in vertebrates limit the understanding of in vivo functions of integrins in glia. Drosophila melanogaster is an excellent model for genetic analysis of the nervous system development. glia.  In this dissertation, I investigated integrin function in Drosophila  Integrins are expressed by glia in both the central and peripheral nervous systems  at larval stages, where they form complexes with Talin and Integrin-Linked-Kinase (ILK). I found that integrin complexes were localized to different glia layers in the larval peripheral nerve and optic stalk.  By using MARCM and RNA interference techniques, I  found that integrins are required for multiple developmental events in individual and populated glia.  In the peripheral nerve, integrins are important for glial ensheathment.  When integrins were removed, perineurial glia failed to initiate or maintain their wrapping around the nerve and wrapping glia failed to send out numerous membrane processes between axons.  In the optic stalk, integrins were necessary for glial migration, ii  deposition and barrier formation. eye disc.  Removal of integrins impaired glial migration into the  Moreover, perineurial glia tended to aggregate at the anterior half and form  multiple layers, and carpet glia failed to form organized septate junctions along the optic stalk.  These glial defects resulted in photoreceptor axonal stalling in the eye disc and  optic stalk, and mis-targeting in the brain.  My work suggests that integrins are  important for different aspects of Drosophila glial development and reveals a new glial function in helping photoreceptor axons through the optic stalk.  Integrin distribution  implicated that integrins may mediate glia-glia or glia-neuron interactions through ECM and non-ECM ligands.  iii  Preface A version of chapter 2 has been published in: Xie, X and Auld, VJ. Integrins are necessary for the development and maintenance of the glial layers in the Drosophila peripheral nerve. Development. 2011 Sep; 138(17): 3812-22. I conducted all of the experiments and data analysis.  Dr. Vanessa Auld and I conceived experiments and  methods of analysis, and wrote the manuscript.  Check the first pages of this chapter to see footnotes with similar information.  iv  Table of contents Abstract........................................................................................................................... ii Preface ........................................................................................................................... iv Table of contents ............................................................................................................. v List of tables ................................................................................................................... x List of figures ................................................................................................................. xi List of abbreviations ..................................................................................................... xiv Acknowledgements ..................................................................................................... xvii Dedication .................................................................................................................. xviii Chapter 1 General introduction ..................................................................................... 1 1.1  Vertebrate glia ............................................................................................ 1  1.1.1  Glia development and functions in the optic nerve .............................. 2  1.1.2  Schwann cell development and function in the peripheral nerve .......... 7  1.2  Drosophila glia ......................................................................................... 12  1.2.1  Glia in the larval peripheral nerve ..................................................... 14  1.2.2  Glia in the optic stalk and eye disc .................................................... 18  1.3  Integrins ................................................................................................... 23  1.3.1  Vertebrate integrins ........................................................................... 26 v  1.3.2  Drosophila integrins .......................................................................... 34  1.3.3  Integrin ligands and binding partners ................................................ 40  1.4  Rationale, hypothesis, and objectives ....................................................... 46  Chapter 2 Integrins are necessary for glial ensheathment formation in Drosophila peripheral nerve ............................................................................................................ 48 2.1  Introduction.............................................................................................. 48  2.2  Methods and materials.............................................................................. 51  2.2.1  Fly strains and genetics ..................................................................... 51  2.2.2  Immunohistochemistry...................................................................... 51  2.2.3  Imaging analysis ............................................................................... 52  2.3  Results ..................................................................................................... 53  2.3.1 2.3.2 nerve  Glial layers in the Drosophila larval peripheral nerve ........................ 53 Specific integrins are located in the glia cells of the larval peripheral ......................................................................................................... 56  2.3.3  The beta integrin subunit is expressed in all glial cell layers .............. 60  2.3.4  Differential glial expression of integrin alpha subunits ...................... 63  2.3.5  Integrin function is necessary in larval peripheral glia ....................... 67  2.3.6  Integrin function is necessary for perineurial glia wrapping ............... 75  2.3.7  The neural lamella is required by the perineurial glia ........................ 81 vi  2.3.8 2.4  Integrin function is necessary for wrapping glia ensheathment .......... 87  Discussion ................................................................................................ 91  2.4.1  The integrin complex in peripheral glia ............................................. 91  2.4.2  Integrin-ECM interactions are required for PG function .................... 92  2.4.3  βPS integrin is required to maintain wrapping glia ensheathment of the  peripheral axons ..................................................................................................... 93 2.4.4  Summary .......................................................................................... 95  Chapter 3 Integrins are required for glial and neuronal development in Drosophila visual system ................................................................................................................. 96 3.1  Introduction.............................................................................................. 96  3.2  Methods and materials.............................................................................. 99  3.2.1  Fly strains and genetics ..................................................................... 99  3.2.2  Immunohistochemistry and imaging analysis .................................. 100  3.2.3  Statistics ......................................................................................... 101  3.3  Results ................................................................................................... 102 3.3.1  stalk  ....................................................................................................... 102 3.3.2  changes  Integrins are expressed in both populations of glia in the larval optic  Loss of βPS integrin in glial cells causes optic stalk morphological ....................................................................................................... 106 vii  3.3.3 optic stalk  βPS integrin is required for glial organization and migration in the  ....................................................................................................... 112  3.3.4  Loss of βPS integrin in glial cells causes axonal migration defects .. 117  3.3.5  Extra glia within the OS does not block axon outgrowth ................. 124  3.3.6  Aberrant glia deposition is observed in the thicker optic stalk ......... 128  3.3.7  Loss of integrins compromises carpet glial barrier in the optic stalk 129  3.4  Discussion .............................................................................................. 135  3.4.1  Integrins play multiple roles in the glia in the optic stalk and eye disc ... ....................................................................................................... 135  3.4.2  Glial organization is important for axonal outgrowth in the optic stalk .. ....................................................................................................... 137  Chapter 4 General conclusion ................................................................................... 140 4.1  Summary of findings .............................................................................. 140  4.2  Significance ........................................................................................... 141 4.2.1  development 4.2.2 4.2.3 development  Insights on using Drosophila to study integrin functions in the glial ............................................................................................. 141 Implications of integrin mediated glia-glia interactions ................... 143 Contribution to the studies of coordination between glial and neuronal .................................................................................................. 144 viii  4.3  Future directions .................................................................................... 145  4.3.1  Identification of glial integrin ligands .............................................. 145  4.3.2  Identification of integrin downstream signalling molecules ............. 148  4.3.3  Examination of glial-glial and glial-neuron interactions .................. 149  Bibliography ............................................................................................................... 152 Appendix A Integrins are required for glia proliferation in the optic stalk ................. 170 A.1 Integrin RNAi in glia gives a thinner optic stalk phenotype......................... 170 A.2 Thinner optic stalk is due to decreased glial proliferation ............................ 174 A.3 Discussion .................................................................................................. 180 Appendix B Semaphorin 1a expression in glia of the eye disc and optic stalk ........... 182 B.1 Correlation between axonal stalling and abnormal glia deposition and differentiation .......................................................................................................... 182  ix  List of tables Table 1: Drosophila integrins. focal adhesion complex components, ECM ligands and their vertebrate homologs. ............................................................................................. 36 Table 2: Summary of markers for ECM and glial layers in the peripheral nerve. ............ 55 Table 3: Integrin and talin RNAi are expressed by 46F-Gal4. ........................................ 80 Table 4. Summary of glial and neuronal phenotypes. ................................................... 111 Table 5. Quantification of glia distribution in the eye disc and optic stalk. ................... 116  x  List of figures Figure 1: Confocal images of rat optic nerve glia ............................................................. 3 Figure 2: Mature Schwann cells as they appear in a transverse section from the sciatic nerve of an adult rat. ........................................................................................................ 9 Figure 3: Glial layers of the larval peripheral nerve ....................................................... 16 Figure 4: Regulation of glial migration and differentiation through FGF-signaling. ....... 22 Figure 5: Sequence relationships of invertebrate and vertebrate integrin subunits. ......... 25 Figure 6: Integrin heterodimer composition. .................................................................. 28 Figure 7: Differential distribution of ECM receptors during SC development. ............... 33 Figure 8: Axons are surrounded by three glial layers in the larval peripheral nerve. ....... 54 Figure 9: Integrins and Integrin-Linked-Kinase (ILK) are expressed in the larval peripheral nerve. ............................................................................................................ 58 Figure 10: Talin forms adhesion complexes with βPS integrin in the larval peripheral nerve. ............................................................................................................................ 59 Figure 11: βPS integrin is expressed in the peripheral glia layers. .................................. 62 Figure 12: αPS2 and αPS3 integrins are expressed in different glial layers of the larval peripheral nerve. ............................................................................................................ 66 Figure 13: MARCM analysis of βPS mutant in the larval peripheral nerve. ................... 70 Figure 14: Expression of integrin-RNAi in glial cells of the peripheral nerve. ................ 74 xi  Figure 15: Expression of βPS-RNAi and talin-RNAi in the perineurial glia ................... 79 Figure 16: Expression of βPS-RNAi in glial cells does not change the neural lamella in peripheral nerves. .......................................................................................................... 83 Figure 17: Overexpression of Mmp2 degrades the neural lamella and causes perineurial glial wrapping defects.................................................................................................... 85 Figure 18: Overexpression of Mmp2 in perineurial glia changes the morphology of the central nervous system (CNS). ...................................................................................... 86 Figure 19: Expression of βPS-RNAi in the wrapping glia. ............................................. 90 Figure 20: βPS integrin and Talin expression in glial cells in the eye disc and optic stalk .................................................................................................................................... 105 Figure 21.  Knock down of integrin and OS morphological changes in early 3rd instar.  .................................................................................................................................... 108 Figure 22. OS morphological changes at wandering 3rd instar. ..................................... 110 Figure 23. Knockdown of integrin and talin compromised glial migration and organization. ................................................................................................................ 115 Figure 24: Photoreceptor axon migration defects observed with integrin loss in glia. ... 121 Figure 25. Photoreceptor axon migration defects in alpha integrin and talin RNAi. ...... 122 Figure 26. Photoreceptor axon targeting defects in RNAi adults. ................................. 123 Figure 27. Axons stall near PG displaced into the center of the optic stalk. .................. 127 Figure 28. Carpet glia defects observed with integrin knockdown. ............................... 132 xii  Figure 29. Model of integrin disruption of glial organization in the OS. ....................... 134  xiii  List of abbreviations BBB  Blood brain barrier  BMP  Bone morphogenetic protein  BNB  Blood nerve barrier  CD8  Cluster of differentiation 8  CG  Carpet glia  CNS  Central nervous system  CNTF  Ciliary-neurotrophic factor  DIC  Differential interference contrast  dpp  Decapentaplegic  ECM  Extracellular matrix  ED  Eye disc  FAK  Focal adhesion kinase  FGF  Fibroblast growth factor  FGFR  FGF receptor  FLP  Flippase  FRAP  Florescent recovery after photobleaching  FRT  Flippase recognition target  gcm  Glial cells missing  GFP  Green fluorescent protein  Gli  Gliotactin  GMR  Glass Multimer Reporter  HRP  Horseradish peroxidase  hs  Heat shock xiv  if  Inflated  IGF-1  Insulin-like growth factor 1  ILK  Integrin-linked kinase  K+  Potassium  kDa  Kilodalton  LIF  Leukaemia inhibitor factor  MARCM  Mosaic analysis with a repressible cell marker  mew  Multiple edematous wings  MF  Morphogenetic furrow  MMP  Matrix metalloproteinase  MTJ  Myotendinous junction  mys  Myospheroid  NA  Numerical aperture  NCAM  Neural cell adhesion molecule  NCC  Neural crest cell  NL  Neural lamella  nls  Nuclear localization signal  Nrg1  Neuregulin 1  nrv2  Nervana 2  NrxIV  Neurexin IV  NT-3  Neurotrophin-3  OPC  Oligodendrocyte precursor cell  OS  Optic stalk  PDGF  Platelet-derived growth factor  PG  Perineurial glia  PH3  Phosphorylated Histone-3 xv  PNS  Peripheral nervous system  PS  Position specific  repo  Reversed polarity  RFP  Red fluorescent protein  RGC  Retinal ganglion cell  RGD  Arginine-Glycine-Aspartic acid  RNA  Ribonucleic acid  RNAi  RNA interference  SC  Schwann cells  scab  Scab  SCP  Schwann cell precursor  Sema1a  Semaphorin 1a  sHH  Sonic Hedgehog  SJ  Septate junction  SPG  Subperineurial glia  TEM  Transmission electron micrograph  ts  Temperature sensitive  UAS  Upstream activation sequence  VEGF-C  Vascular endothelial growth factor C  wb  Wing blister  WG  Wrapping glia  WT  Wild type  xvi  Acknowledgements I would not complete this dissertation without the support and cooperation from many people through various means. Foremost, it is my honour to express my sincere gratitude to my devoted supervisor, Dr. Vanessa J. Auld. Her guidance, patience and encouragement enabled me to develop an interest of the field all the way from the very beginning. Throughout my PhD study, I was guided by her with profound knowledge and invaluable ideas. Her support and kindness are heartily appreciated. I would like to thank my committee members Dr. Douglas Allan, Dr. Jane Roskams, and Dr. Linda Matsuuchi for their priceless suggestions and discussions over the years, including their input on writing this dissertation. I would also like to thank all previous and current members of the Auld lab, especially Dr. Murry Glibert, Dr. Patrick Cafferty and Dr. Mojgan Padash for their support and friendship. At last but not least, I would like to thank the entire fly community for their generosity on sharing various reagents, which make my life much easier. .  xvii  Dedication  This dissertation is dedicated to: My parents for raising me with love, patience and tolerance; My dear wife Jing for supporting me with constant love and encouragement; My daughter Jenny for bringing me infinite happiness.  xviii  Chapter 1  General introduction  1.1 Vertebrate glia Glia and neurons are the two major classes of cells in the nervous system.  Glia  were given the name as ‘nerve glue’ when they were first discovered by Rudolph Virchow at 1846 and were viewed as merely connective tissues in the brain [1].  Now  glia are well known for providing neurons with structural, metabolic and trophic supports and more recent evidence suggests that glia are active players in many major aspects of nervous system development and functioning. Glia are generally assorted as macroglia and microglia in the vertebrate nervous system.  Macroglia and microglia originate from different germ layers during the  embryogenesis. neurons.  Macroglia are derived from the ectodermal neuroepithelium with  Microglia however are mesodermal cells and not always counted as ‘true’ glia.  Microglia are immune cells in the central nervous system (CNS) which survey and respond to pathological changes in the brain [2]. Macroglia constitute the majority of the glial population in the nervous system.  The main macroglia include  oligodendrocytes and astrocytes in the central nervous system (CNS) and Schwann cells in the peripheral nervous system (PNS) [3].  Oligodendrocytes and Schwann cells are  the axon ensheathing glia, which wrap axons with or without myelin formation. Astrocytes are broadly distributed in the CNS and have the most diverse morphologies 1  and functions in all glial cell types.  Other specialized macroglia are found in restricted  regions, such as: Müller cells in the retina [4], ependymal cells in the ventricle [5], olfactory ensheathing cell in the primary olfactory system [6], satellite glia in the peripheral ganglia [7] and enteric glia in the enteric nervous system [8]. In the following sections, the optic nerve and peripheral nerve will be used as samples to explain the basic principles and concepts of CNS and PNS glial development and functions, respectively.  1.1.1  Glia development and functions in the optic nerve  The optic nerve is the second cranial nerve, in which the retinal ganglion cell (RGC) axons travel from the eye into the brain [9]. will give rise to the eye in later stages.  RGCs are formed in the optic cup which  Prior to RGC differentiation, the optic cup is  connected to the brain by a neuroepithelial structure called the optic stalk.  The optic  stalk is transformed to the optic nerve when the RGC axons grow out of the retina.  The  two major CNS glial types are oligodendrocytes and astrocytes (Fig. 1) and are both found in the optic nerve.  Since there are no neuron cell bodies in the optic nerve, this  relatively simple organization makes the developing optic nerve a good model to study the role of glia during CNS development.  2  Figure 1: Confocal images of rat optic nerve glia (A) Oligodendrocytes have 5–50 parallel processes labeled with intracellular-dye and show the dye-filled inner and outer tongues of the internodal myelin sheath.  (B)  Astrocytes extend thick processes perpendicular and parallel to the axons and terminate in bulbous swellings or end feet at the pia and on blood vessels.  Scale bar, 50 µm.  Reprinted by permission from Macmillan Publishers Ltd: [Developmental Biology] [10], copyright (1984).  3 Origin and functions of astrocytes Two distinct lineages of astrocytes have been found in the optic nerve.  Type-1  astrocytes are derived locally from the optic stalk neuroepithelium during the embryogenesis [11, 12].  Type-2 astrocytes are generated from the oligodendrocyte  precursor cells (O-2A precursors) by in vitro studies [13] but little evidence supports the existence of type-2 astrocytes in vivo [14].  Astrocyte development in the optic nerve  depends on both non-neuronal cells and RGC axons [15-18], which provide signals such as: leukaemia inhibitor factor (LIF), fibroblast growth factor 2 (FGF2), neuregulin 1 (Nrg1), sonic hedgehog (sHH) and bone morphogenetic protein 7 (BMP7) [16, 19-23]. Astrocytes are heterogeneous in their structure and function throughout the CNS. The grey matter astrocytes (protoplasmic astrocytes) are greatly different from the white matter astrocytes (fibrous astrocytes) in their morphological, physiological and biochemical properties [24]. the same.  Even astrocytes from similar locations are not necessarily  For example, the optic nerve astrocytes can be divided into three  morphological classes for having either transverse, random, or longitudinal process orientation [25]. The diversity of astrocyte morphology reflects their diverse functions, which include (but not limited to) providing structural, metabolic and trophic support of neurons, regulating extracellular ion concentrations, uptake of and recycling of neurotransmitters, promoting angiogenesis and blood-brain barrier formation, regulating synaptogenesis and synaptic activities [26]. 4  Some of these functions were first  discovered in optic nerve astrocytes.  For example, potassium ions (K+) are released  from nodes of Ranvier during axonal transmission, and consistent release of K+ can increase extracellular K+ concentration and cause neuron depolarization [27].  The optic  stalk astrocytes are observed to send processes to contact nodes of Ranvier and protect neurons from hyperexcitability by removing K+ from the extracellular space [28, 29]. Origin and functions of oligodendrocytes Unlike astrocytes, oligodendrocytes are produced postnatally [12, 13] from oligodendrocyte precursor cells (OPCs), which originate from embryonic ventral midline brain regions [30].  Optic nerve OPCs undergo extensive migration through the optic  chiasm (a structure where the bilateral optic nerves join together) to populate the optic nerve [31, 32].  Long distance glial migration is a common phenomenon in many CNS  regions, where glial precursors are mostly specified in limited areas and disperse into broad territories in the brain and spinal cord [33, 34].  OPC migration and RGC axon  outgrowth happen around the same time but at opposite directions in the optic stalk/nerve. LacZ labeled OPCs have been shown to start their migration from the chiasm at around mouse embryonic day 14.5 (E14.5) and reach the optic nerve head (the lateral end of the optic nerve) at around E17.5 [32].  Directed OPC migration relies on secreted  chemorepellent guidance cues Netrin-1 and Semaphorin 3a [31, 32] as well as extracellular matrix protein tenascin-C [35].  OPCs cease their migration before entering 5  the optic nerve head (a structure between the optic nerve and eye).  The migration  termination signal might be transiently expressed Netrin-1 in the optic nerve head [36]. After moving into the optic stalk/nerve, OPCs undergo cell proliferation and produce differentiated oligodendrocytes that match the number of RGC axons in the optic nerve.  RGC axons have been shown to have essential roles in oligodendrocyte  development.  First, RGC axons promote OPC proliferation and oligodendrocyte  survival [37, 38].  Axon derived neuregulin and the notch ligand, Jagged 1, are  important in these processes [39, 40].  Second, RGC axons regulate oligodendrocyte  maturation and myelination by producing soluble FGFs [41, 42] and thyroid hormone [43], and cell surface molecules, such as L1, neural cell adhesion molecule (NCAM) and N-cadherin [44-46].  In addition to the factors from RGC axons, astrocyte derived  trophic factors also play important roles in oligodendrocyte development.  These factors  include (but are not limited to) platelet-derived growth factor (PDGF), insulin-like growth factor 1 (IGF-1), neurotrophin-3 (NT-3), ciliary-neurotrophic factor (CNTF) and leukaemia inhibitory factor (LIF) and vascular endothelial growth factor C (VEGF-C) [47-50]. The primary function of oligodendrocytes is to insulate axons with myelin sheaths, a multiple-layered wrapping around axons necessary for fast conduction including the transmission of visual information along the optic tracts. ensheath multiple axons.  Each oligodendrocyte can  In the rat optic nerve, each oligodendrocyte appears to 6  ensheath 20-30 axons with 150-200 µm long parallel internodal myelin processes [51, 52]. Myelination is not the only function of oligodendrocytes; they are also important for the nodes of Ranvier formation [53-55] and maintenance of axon integrity all through the CNS [56-58].  In summary, the studies in the optic nerve suggests that CNS glial development including glial migration, proliferation, survival and maturation, are highly regulated by both intrinsic cellular programs and extrinsic environmental cues.  The glia-neuron and  glia-glia interactions have critical roles in these developmental processes.  1.1.2  Schwann cell development and function in the peripheral nerve  Schwann cells (SCs) are the major glial type that envelope the PNS axons with both myelinating and non-myelinating forms (Fig. 2). Both myelinating and non-myelinating SCs are derived from the neural crest cells [59].  The myelinating or non-myelinating  fate is determined by the axons that SCs associate with.  Myelinating SCs form in  conjunction with large diameter (>1µm) axons and each SC forms myelin around a single axon.  Non-myelinating SCs form Remak fibres around small diameter axons (C-fibers  or nociceptive axons) in which each SC embeds several axons in their membrane protrusions [60, 61]. The development from neural crest cells to functional SCs involves the generation of 7  two sequential intermediates: Schwann cell precursors (SCPs) and immature SCs [59]. Neural crest cells are generated during the neurulation and then segregate from the dorsal tip of the neural tube.  They migrate peripherally and give rise to various cell types  including melanocytes, PNS neurons and glia.  SCPs associate with axon bundles in  early stage nerves (mouse E12-13 and rat E14-15) and generate immature SCs in later stages (mouse E13-15 and rat E15-17).  Immature SCs are present in the embryonic  nerves until birth and their postnatal fate is determined by the axons that they associate with.  8  Figure 2: Mature Schwann cells as they appear in a transverse section from the sciatic nerve of an adult rat. (A) A myelinating Schwann cell forms a compact, multilayered sheath (M) around a single large diameter axon (Ax). Parts of other myelinated fibers are present. (B) A cross-section through the nucleus of a non-myelinating Schwann cell (N-M) that ensheaths 13 axons (for example, A), each lying in a separate trough in the cell surface. Non-myelinating cells can also ensheath axons singly as shown for A*. Myelin sheaths (M) of neighboring axons are visible and the axon–Schwann-cell units are separated by collagen-rich (C) extracellular spaces. Scale bar, 0.5 μm. Reprinted from Trends in Neuroscience, 1999, by permission [62].  9  Similar to CNS glia, the development of peripheral nerve glia is also regulated by complex intrinsic and extrinsic signals.  Among these factors, axonal derived neuregulin  1 (Nrg1) through its ErbB receptors seems to have instructive roles in multiple stages of SC development [63].  Nrg1 allows or promotes glial cell fate from the neural crest cells  [64] and acts as a trophic factor to maintain SCP and immature SC survival and proliferation [65-69].  In later stages, Nrg1 is required for SC migration and  differentiation prior to myelination [70, 71]. promotes myelination [72-74].  At the last stage of SC development, Nrg1  Different to CNS oligodendrocytes, an ECM basement  membrane is deposited around each immature and mature SC and acts as another source of regulatory factors.  The roles of ECM-SC interactions will be discussed in section  1.3.1. As the major glial cells, SCs and their progenitors have many important functions in the peripheral nerve.  Similar to CNS oligodendrocytes, a primary function of  myelinating SCs is to promote fast transduction of the action potential along axons.  In  developing nerves, SCPs and immature SCs not only receive trophic signals from axons but also provide essential survival factors to developing neurons [75, 76]. SCs also have morphogenetic functions during peripheral nerve formation.  Developing Boundary  cap cells or the SCP that are derived from them help to keep motor neurons in the spinal cord [77].  SCP and immature SC are required for peripheral nerve fasciculation [78, 79].  SC derived sonic hedgehog (sHH) signals regulate the connective tissue formation in the 10  peripheral nerve [80]. In summary, glia are critical players in both the central and peripheral nervous systems.  They support neurons in multiple ways and are involved in regulating neurons  and other cell types development in the nervous system.  Reciprocally glial development  is greatly influenced by extrinsic factors which are present by neurons, other glia and the ECM.  11  1.2 Drosophila glia During evolution the proportion of glia generally increases in the nervous system. For example, glia occupy about 25% of cells in the Drosophila brain and vastly outnumber neurons in the human brain [81, 82].  Although invertebrates have less  complex nervous systems and less abundant glial cells, diverse glia subtypes are found in the invertebrate nervous systems and share many structural, morphological, molecular and functional similarities as vertebrate glia [83].  Therefore invertebrate organisms,  including fruit flies have been studied to answer fundamental questions about glial development and glia-neuron interactions. By using various genetic tools, comparable glial subtypes are observed in both the CNS and PNS in Drosophila as in vertebrates (reviewed by Freeman [83] and Hartenstein [84]).  For instance, three major classes of glia have been classified in the larval brain  and ventral nerve cord based largely on their morphologies and positions: the surface glia, the neuropil glia, and the cortex glia [83, 85]. Surface glia surround the Drosophila CNS with a two layered sheath of flat cells that include outer perineurial glia (PG) and inner subperineurial glia (SPG) [86].  Cortex glia span the thickness of the cortex and  send out processes around neuron bodies, which are proposed to supply gas and nutrient to the neurons, acting as the vertebrate astrocyte [86].  Neuropil glia are localized to the  inner neuropile regions where they send out membrane processes around axonal fascicules and mimic the vertebrate oligodendrocyte [86, 87]. 12  Drosophila glia are important for multiple aspects of nervous system development. For example, Drosophila glia can guide and facilitate axon outgrowth and termination. In embryos, glia provide particular guidance cues to help axons cross or avoid the CNS midline [88-91]. Sensory axons require glia located at the PNS/CNS boundary to enter the CNS [92, 93].  In the developing visual system, glia act as intermediate targets for  photoreceptor axons to terminate properly in the optic lobe [94] and constitute compartment boundary for different visual centers [95].  During metamorphosis, glia  facilitate axon pruning in the developing mushroom body, the memory center in the Drosophila CNS [96, 97]. Other than their developmental roles, Drosophila glia are also essential for nervous system function.  Drosophila glia express glutamate transporters and glutamine  synthetase, which function to maintain neurotransmitter homeostasis [98-106]. Drosophila surface glia (or subperineurial glia) constitute the blood-brain barrier by forming septate junctions between neighboring glia to protect internal neurons and axons from pathogens and the high concentration of potassium ions in the hemolymph [107-109].  Moreover, glia can respond to neuronal injuries in adult flies and clear  neuronal debris when axons are severed [110]. The development of glia has been well characterized in Drosophila. In Drosophila embryos, transient activation of a transcription factor, glia cells missing (gcm) is both necessary and sufficient for specifying the glial fate (except the CNS midline glia) 13  [111-114].  Mammalian gcm homologs (Gcm1 and Gcm2) however are mostly found in  non-neural tissues, suggesting that glia specification is controlled by different molecular mechanisms in the Drosophila and vertebrate nervous systems.  Nevertheless, later  aspects of glial development (i.e. proliferation, migration, differentiation, interaction with neurons and extracellular environment and neuronal ensheathment) share many similarities at the molecular level [83].  Among all of these diverse Drosophila glial  subtypes, two groups of glia have been shown to be good models for studying different aspects of glial development and functions and will be reviewed in the following sections.  1.2.1  Glia in the larval peripheral nerve  The larval peripheral nerve connects the CNS and peripheral sensory organs and muscles.  Peripheral nerves contain both motor and sensory neuron axons wrapped by  distinct layers of glia (Fig. 22).  At late embryonic stages, the axon fascicules are  initially ensheathed by peripheral glia, which are either born at the edge of the CNS and migrate peripherally out along axons or are born in the periphery and migrate towards the CNS [93, 115].  Three morphological and molecular distinct glial layers around the  axons start to form in the 1st instar larvae and become fully mature by the 3rd instar larvae. These glia are the innermost wrapping glia (WG), intermediate subperineurial glia (SPG) and the outermost perineurial glia (PG).  The WG and SPG are most likely differentiated 14  from the embryonic peripheral glia and the outermost PG may originate from the mesoderm [116].  It is still unclear how the embryonic glial sheath transits to the  three-layered larval glial sheath in the peripheral nerve because of the lack of direct observation of the transition in the late embryonic and 1st instar larval nerve.  Other than  different origins, these glia also behave differently during later development stages. Both the WG and SPG become post-mitotic and polyploid and significantly expand their cellular membranes to match axonal growth during the larval stages [92, 117]. this stage the smaller PG actively proliferate to cover the expanding nerves [118].  15  During  Figure 3: Glial layers of the larval peripheral nerve TEM cross-section of a peripheral nerve from first instar (A) and late third instar larvae (B and B′).  Surface glia comprise an outer perineurial glial (png; light blue) layer and  an inner subperineurial layer (spg; dark blue). Axons (ax) are surrounded by wrapping glia (wg, red).  Note that perineurial glia has not yet expanded around the entire nerve in  early larva, as opposed to 3rd instar larvae where they completely surround the nerve (B); similarly, wrapping glia form processes around individual axons only in late 3rd instar larvae.  B′ shows a magnified view of the boxed area in B, illustrating the auto-septate  junction (sj) of subperineurial glia. Arrows in (B) point at protrusions of subperineurial glia contacting axons, a phenomenon not observed in subperineurial glia of CNS. 1 μm.  Reprinted from Glia, 2011, by permission [84].  16  Bar =  The three glial layers have distinct morphologies and, presumably, functions.  Two or  three WG are usually found in each peripheral nerve and their nuclei are evenly distributed along the nerve.  Numerous membrane processes are sent out from the WG  body to both directions and can cover a distance of hundreds of microns.  Although no  specific function has been confirmed for the WG, they are postulated to resemble the vertebrate non-myelinating Schwann cells and facilitate bi-directional axonal conduction by separating motor and sensory axons [109]. Similar to the WG, the SPG are also large cells and their nuclei are spaced distant to each other.  However, each SPG spreads out single layer of membrane and forms a tube  around the axons and WG.  An essential role of the embryonic peripheral glia and later  SPG is to form a brain-nerve barrier (BNB), which insulates axons from the high concentration of potassium ion in the hemolymph [107, 109].  Septate junctions (SJs),  an electron dense ladder-like structure, were initially identified between embryonic peripheral glial membranes [108, 119] and at later stages between SPG membranes [109]. The SJs are critical for the barrier formation.  Loss of SJ proteins, such as Neurexin IV  (NrxIV), Contactin, or Neuroglian blocks potential conduction and leads to paralysis [108, 119, 120]. The perineurial glia (PG) are the most abundant glial subtype in the larval peripheral nerve.  The PG number varies between each segment and can rise to about 100 in the  longest nerves (A8) at the wandering 3rd instar stage. 17  The PG are spindle or squamous  like cells that are tiled with each other to form a single layer surrounding the underlying SPG.  The function of the PG is not completely understood.  They are suggested to  contribute to a septate junction independent barrier, which along with the neural lamella (NL) is selective against large macromolecules or pathogens [109]. The NL is an ECM layer which surrounds the peripheral nerve and consists of common ECM components such as Collagen IV [109].  A thin NL is first observed  around the peripheral glia at embryonic stages and a thick ECM layer is deposited on the PG during the larval stages [119].  Hemocytes have been shown to be critical for the NL  formation and are thought to contribute to the deposition of ECM components during embryogenesis [109, 121].  The ECM is capable of being a substratum for cell adhesion  and migration but the role of the NL in Drosophila glial development is still unclear.  1.2.2  Glia in the optic stalk and eye disc  The adult fruit fly compound eye originates from a larval tissue called the eye imaginal disc or eye disc.  In the retina, photoreceptor cells are organized into ~750  repeating units, named the ommatidia.  Ommatidial differentiation begins and  progresses in a posterior-to-anterior orientation in the middle and late 3rd instar larval eye disc [122].  The front edge of the wave of differentiation is marked by an indentation of  the disc epithelium, called the morphogenetic furrow (MF), which propagates towards the anterior end of the eye disc accompanying ommatidial formation. 18  Posterior to the MF,  nascent photoreceptor cells send out axons basally and posteriorly in the eye disc, through the optic stalk, and then into the brain.  Each ommatidium contains eight  photoreceptor cells (R1-8) whose growth cones target to different regions in the optic lobe [123].  R1-6 axons innervate the lamina, whereas R7 and R8 terminate in the  deeper medulla region.  The axonal outgrowth into the brain requires physical contact  with, and support from, a population of glia that originate in the optic stalk [124]. The optic stalk (OS) is a tubular structure constructed by glial cells and mediates the connection between the eye disc and the brain in the Drosophila larvae. the optic stalk glia is unclear.  The origin of  They might derive from the CNS and migrate along the  larval photoreceptor Bolwig’ nerve through a process similar to the migration of the peripheral glia along the segmental nerve at the late embryonic stages [125].  However,  these glia have not been visualized until the 2nd larval instar when 6-25 glial cells have been observed in the OS [124]. At the early 3rd instar before the entry of the R-cell axons, two major classes of glia are clearly seen around the Bolwig’s nerve in the optic stalk: inner carpet glia (CG) and the outer perineurial glia (PG) (Fig. 4A). through larval stages.  Two CG are constantly found in the OS  They are most likely specialized SPG because SJs are found  between the two CG [126]. Therefore the CG likely create a permeable barrier around the OS.  The CG are different from other SPG in the CNS and peripheral nerves with  regards to their polarity along the anterior-posterior axis. 19  After the R-cells form and  their axons enter the optic stalk, the CG anterior membranes expand and form a flat mesh like structure in the eye disc while the CG posterior membranes maintain a tubular structure in the OS. The perineurial glia (PG) in the OS are named for the morphological, molecular and structural similarities to other PG in the CNS and peripheral nerve.  They are abundant  and proliferate through the larval stages.  Collaboratively the squamous PG form a  single layer around the outside of the CG.  The PG in the OS also have distinct motility  and differentiation properties.  They were also named retinal basal glia or subretinal glia  in previous literature for their migration pattern along the basal surface of the eye disc [124, 127]. The timing of PG migration and CG spreading is critically controlled to correspond to the same time as photoreceptor differentiation.  Therefore PG migration is  initially inhibited by the eye disc epithelium through a Casein kinase called gilgamesh, the eye specification transcription factors (eyeless, eye absent, sine oculis) and secreted hedgehog protein [128].  OS glial migration is specifically triggered by the  differentiation of photoreceptor cells through a mechanism which is independent of photoreceptor axons [124] but involves the hedgehog, decapentaplegic and fibroblast growth factor (FGF) signaling pathways [128-130] (Fig. 4B). also have roles in glial proliferation in the eye disc.  These signaling pathways  Constant PG migration and  proliferation increase glia number in the eye disc, which coordinates and matches the numbers of glia to the gradual increase in ommatidia throughout late 3rd larval instar [126, 20  127] (Fig. 4C). FGF signaling regulates PG proliferation, migration and differentiation through a switch between two FGF ligands [130].  One FGF ligand Pyramus is present in the CG  and modulates PG proliferation and movement.  A second FGF ligand Thisbe is  produced by photoreceptor neurons, and can be detected only by pioneer migrating PG when they migrate close to the MF and beyond the boundary of the CG.  The absence of  the intervening layer of CG allows the axonaly produced Thisbe to bind to the FGF receptor (heartless) on the PG surface and initiate the fate change from PG to WG [130]. The WG are easily identified in the optic stalk and eye disc for their long processes along photoreceptor axons toward the brain and short projections into the apical regions of the eye disc.  The eight axons from individual ommatidia are always bundled together  on their way to their destinations in the brain [130].  WG processes lie between these  axon bundles and are thought to help maintain this separation.  The WG in the eye disc  send out simple protrusions to ensheath a limited number of photoreceptor axonal bundles, unlike the WG in the peripheral nerve which extend numerous processes wrapping around all the axons in the nerve.  The WG ensheathment formation is also  controlled by the FGF signaling pathway.  Down-regulation or increase of FGF  signaling can cause hypo- or hyper-wrapping phenotypes in the eye disc and optic stalk, respectively [130].  However, it is unclear whether these wrapping phenotypes are  generated through changes of WG number or wrapping ability of individual WG. 21  Figure 4: Regulation of glial migration and differentiation through FGF-signaling. The glia migration onto the eye disc (ED) is depicted in three successive developmental stages. (A) In early larval stages a still elusive signal from the ED suppresses carpet glia (CG) growth onto the ED. (B) Upon initiation of morphogenetic furrow (mf) formation, this block is removed and the CG extends on the ED. Decapentaplegic (Dpp) expression at the furrow or FGF expression anterior to the furrow (indicated by +) may attract the CG. First contact of the perineurial glia (pg) with nascent photoreceptor (pr) axons triggers the differentiation of the wrapping glia. (C) As the mf progresses over the eye disc, the CG follows the furrow and successive cycles of glial differentiation are induced.  Reprinted from Fly (Austin), 2010, by permission [125]. 22  1.3 Integrins Glial cells, like many other cell types in multicellular organisms, rely on surface cell adhesion molecules (CAMs) to attach to each other and to the extracellular matrix (ECM). CAMs are not only necessary to hold cells together, but also are involved in various cell activities, including: cell proliferation, migration, differentiation, polarization, growth, death and survival.  Various types of CAMs, including cadherins, immunoglobin  superfamily and integrins are present in animals ranging from worms and insects to vertebrates in which they are expressed by certain cell types, locate to specific subcellular domains and mediate particular cell adhesions [131]. Integrins, one family of cell adhesion receptors, have been shown to play important roles in mediating both cell-ECM and cell-cell interactions.  Integrins are heterodimeric  transmembrane proteins composed of one alpha and one beta subunit.  Both subunits are  type I transmembrane glycoproteins that contain a larger extracellular portion (about 700 amino acids for alpha and 1000 amino acids for beta subunits) and a shorter intracellular tail (less than 75 amino acids except for the vertebrate β4 subunit) [132].  The  extracellular part of integrin subunits contain multiple domains which generally fold into an elongate stalk and a globular ligand-binding head region [133-137]. The intracellular tail, although small, contains multiple binding motifs that can form complexes with a large number of downstream proteins [138].  Through interactions on both sides of the  membrane, integrins can mediate cells attachment onto specific substrates and transduce 23  extracellular signals into the cells to influence particular aspects of cell activities [139]. As with many other CAMs, the integrins are conserved throughout the animal kingdom (Fig. 5).  At least one alpha and one beta integrin subunit has been cloned from  the most ancient animals, the sponges [140, 141], where the integrins already have important functions in cell-cell adhesion [142]. The nematode Caenorhabditis elegans has two alpha and one beta integrin subunits [143].  The fruit fly Drosophila  melanogaster hosts five alpha and two beta subunits [143].  The integrin family expands  considerably in the vertebrates.  As many as 26 alpha and beta subunits have been  discovered in the human genome [144].  While the vertebrate integrins evolve novel  extracellular ligand binding and intracellular interaction properties and participate in more complex biological activities that do not exist in lower animals, many still preserve the basic compositional, structural and functional characterizations as their ancient homologues.  In the following sections, vertebrate and Drosophila integrins will be  introduced to show the similarities and differences.  24  Figure 5: Sequence relationships of invertebrate and vertebrate integrin subunits. Comparison of Drosophila, C. elegans and human integrin α subunits at the left and β subunits at the right. To help with the comparison, other invertebrate sequences are included from sea urchin and sponge.  To visualize the rates of divergence within the  vertebrates, a Xenopus α and β are also included. with ClustalW (MacVector software).  Alignment and tree were generated  The human integrin α subunits that are  homologous to invertebrate integrin α subunits are divided into two groups of laminin-binding and RGD-binding subunits according to their ligand binding preferences. This figure is adapted from Matrix Biology, 2000, by permission [143].  25  1.3.1  Vertebrate integrins  The vertebrate integrin family includes 18 alpha and 8 beta subunits which form at least 24 different integrin heterodimers (Fig. 6). The integrin alpha subunits can be equally divided into two groups based on their sequence.  The first group of alpha  subunits contains an I or A domain inserted in their extracellular part, which as well as the I-like domain in the integrin β subunits participate in the ligand binding.  I-domain  containing alpha subunits are only found in the vertebrates, suggesting their late onset during integrin family expansion [143].  The second group alpha subunits (lacking the  I-domain) can be further divided into three subgroups.  One subgroup of alpha subunits  generates integrins that recognize RGD (arginine-glycine-aspartic acid) peptide containing ligands.  Another subgroup makes up integrins that bind laminins.  group consists of the rest alpha subunits, α4 and α9.  A third  The RGD-binding and laminin  binding alpha subunits are homologous to invertebrate integrins, suggesting that they arose early during evolution (Fig. 5). Integrins are involved in a large number of biological processes. Some integrins are For example, α4, β2 and β7 integrins are  expressed in restricted regions in the animals.  exclusively present on the surface of the leukocytes, where integrins are important for various phases of the immune responses [145, 146].  Many integrins are broadly  distributed in many tissues and cells types, such as the β1 integrins, which are found in many tissues including skin keratinocytes [147], bone osteoclasts and chondrocytes [148, 26  149], cardiac and skeletal muscle cells [150, 151], and vascular endothelial cells [152]. β1 integrins as well as other integrins are also found in neurons, glia and other cell types within the nervous system, where integrin mediated cell-cell and cell-ECM interactions have been shown to be important for establishing the three dimension architecture and maintaining the normal function of the nervous system [153-155].  In the following  sections, I will introduce the roles of integrins in the development of CNS oligodendrocytes and PNS Schwann cells .  27  Figure 6: Integrin heterodimer composition. Integrin α and β subunits form 24 heterodimers that recognize distinct but overlapping ligands.  Half of the α subunits contain I domains (asterisk). Blue circles label integrin  subunits that have been detected in oligodendrocytes and Schwann cells.  The figure is  adapted and modified from Nature Reviews Drug Discovery, 2003, by permission [156].  28 Integrins in oligodendrocyte and astrocyte development. A role for integrins in oligodendrocytes was first implicated from in vitro assays where a RGD containing peptide suppressed adhesion and differentiation of cultured oligodendrocytes [157-159].  Later studies revealed that multiple integrins, such as αV  integrins are expressed in the oligodendrocyte lineage and their expression is tightly regulated in different development phases [159-161].  In vitro functional studies further  show that different integrins have roles in different aspects of oligodendrocyte development.  For example, αVβ1 integrin is required for migration of oligodendrocyte  progenitors, and α6β1 integrin is necessary for oligodendrocyte survival and myelination [162-167].  The role of integrins in oligodendrocyte survival and myelination were  confirmed using in vivo studies with either a dominant negative form of β1 integrin or conditional knock-out of the β1 integrin gene [168-171]. Consistent with the studies on oligodendrocyte integrins, laminins, a major integrin ligand, are also required for oligodendrocyte development.  Laminin-2 and laminin-5  provided by the astrocytes appear important for integrin mediated oligodendrocyte survival [164].  In addition, axon derived laminin-2 acting through integrins is the signal  for oligodendrocyte survival and myelin formation [163, 172-174] . Binding of integrins to their ligands can be facilitated by other cell surface receptors on the oligodendrocyte.  For example, the PDGFα (platelet derived growth factor alpha)  receptor can associate with and activate two integrin heterodimers in different 29  oligodendrocyte developmental stages.  Activation of αVβ3 integrin stimulates  oligodendrocyte progenitor proliferation [175], and activation of α6β1 promotes oligodendrocyte survival and differentiation [173].  The different integrin signals are  transduced through distinct downstream signalling partners that associate with specific integrins, such as: the Src family kinase Lyn for αVβ3 integrin and the Src family kinase Fyn for α6β1 [176]. Integrins in Schwann cell development. Schwann cell (SC) development starts with the migration of neural crest cells (NCC), precedes through Schwann cell precursors (SCP), immature SCs and promyelinating SCs, and ends up with fully mature myelinating or non-myelinating SCs (Figure 4).  As  opposed to CNS oligodendrocytes, a basement membrane starts to form around immature and mature SCs.  Spontaneous mutations in laminin genes trigger nerve dysmyelinating  neuropathies in both mice and humans [177-182]. In vivo studies confirm that laminin-2 and -8 are present in the basement membrane around SCs in developing peripheral nerves [183, 184], suggesting that laminin receptors such as integrins and dystroglycan are involved in SC development. Multiple integrins have been detected in all different phases of SC development and their expression is tightly regulated [155, 185] (Fig. 7).  The NCCs in vitro express a  large number of integrins, which consist of α1β1, α3β1, α4β1, α5β1, α6β1, α8β1, αvβ1, 30  αvβ3 and a β8 integrin [186, 187]. integrins are observed. development.  In later developmental stages less diversity of  However β1 integrins are consistently found throughout SC  For example, α6β1 integrin is observed all through the later SC lineage.  Interestingly, myelinating and non-myelinating SCs appear to share a number of integrin receptors with the exception of α1β1 integrin, which is only present in non-myelinating SCs. Integrins are critical for different stages of SC development.  In the NCC integrins  are necessary for cell adhesion, migration, survival, and proliferation in vitro [186-190]. Some integrin functions have been confirmed by in vivo studies.  For example, ablation  of β1 integrin in the NCC caused a delay of NCC migration [191].  Knockout of α4 and  α5 integrins showed NCC survival and proliferation defects respectively [192]. For the SCP and immature SC, their migration along axonal tract and radial sorting in between axonal bundles are critical for successful myelin formation.  These two  different glial movements seem to be regulated by different integrins.  When β1  integrins were conditionally ablated, the SC failed to initiate and maintain their processes around axons, suggesting a role of β1 integrin in radial sorting [193].  SC membrane  extension requires rearrangement of actin cytoskeleton, which is mediated by integrin downstream integrin-linked-kinase and Rac1 activation [194, 195].  However, another  direction of SC movement along the axon tracts appears to be independent of β1 integrins [193, 195] and might be regulated by other receptors, such as αvβ8 integrin [196]. 31  Although integrins are also expressed by non-myelinating SCs, their roles in this class of glia are less clear.  Laminin has been shown to participate in non-myelinating SC  development and Remak fibre formation [197]. However, no integrin mutants tested so far have been shown to generate a non-myelinating SC phenotype.  In summary, studies in both oligodendrocytes and Schwann cells show that integrins are required for both CNS and PNS glia and are involved in many aspects of glial development all through glial lineage.  However, many integrin functions have been  suggested by in vitro studies and not completely confirmed by in vivo experiments. is partially due to the difficulties of conducting in vivo assays in mammals.  This  The large  number of vertebrate integrin family members, the potential redundancy and compensation between different subunits adds another level of difficulty.  Thus it seems  necessary to investigate a relatively simple model organism to study the basic functions of integrins in glial cells.  32  Figure 7: Differential distribution of ECM receptors during SC development. NCC express several integrin receptors, but in the following steps of development the distribution of these receptors becomes more restricted.  α6β1 appears in SCP and in  immature SCs, and remains expressed throughout the whole of the SC lineage.  α6β4  starts to be expressed by promyelinating SCs, first in a diffuse fashion, and then polarized is expressed after the first postnatal week.  Whereas  there is no integrin expressed only in myelin-forming SCs, α1β1 integrin is specific for nonmyelin-forming SC. SC, Schwann cell; NCC, Neural crest cells; SCP, Schwann cell precursor. This figure is reprinted from NeuroMolecular Medicine, 2006, by permission [155].  33  1.3.2  Drosophila integrins  The Drosophila Melanogaster has a small integrin family, which includes five alpha (αPS1-αPS5) and two beta (βPS and βν) subunits.  These were historically named PS  proteins for their position specific rather than cell type specific expression pattern in Drosophila wing epithelial cells [198, 199]. their mutant phenotypes.  Integrin genes are named differently for  The αPS1-3 are encoded by genes of multiple edematous wing  (mew), inflated (if) and scab (scb)/volado, respectively, and the βPS subunit is the product of myospheroid (mys) (Table 1). The αPS1 subunit is homologous to vertebrate α3, α6 and α7, while αPS2 is equally similar to vertebrate α5, α8, αV and αIIb (Figure 2) [200]. The sequence similarities also extend to their ligand preference.  αPS1βPS (PS1) integrin is a laminin receptor and  αPS2βPS (PS2) integrin binds to ligands containing the tripeptide sequence RGD [201, 202].  The αPS3-5 subunits are less conserved to the vertebrate integrin subunits [131].  They are closely related and appear to result from recent duplication events.  The  αPS3βPS integrin has been shown to bind to both forms of laminin although it does not appear to be the ortholog of vertebrate laminin binding integrins [203, 204]. Between the two Drosophila integrin beta subunits, βPS has sequence conservation with more than one vertebrate beta subunit including: β1, β2 and β7.  The other  Drosophila integrin beta subunit βν has no obvious vertebrate orthologs (Figure 2) [200]. The βPS has been demonstrated to form heterodimers with αPS1-3 in a broad range of 34  cell and tissue type and deletion of βPS (mys) results in embryonic lethality [205].  The  expression of the second βν subunit is mostly restricted to the developing midgut endoderm and their precursors, and deletion of βν is not lethal [206]. As in vertebrates, integrins are the major ECM receptors in the Drosophila.  They  play important roles in many cell and tissue types from embryos to adults that include mediating stable cellular adhesion [207, 208], regulating cell migration [208, 209], modulating axonal outgrowth [210] and short term memory [211], maintaining the stem cell niche [212, 213].  Among these diverse functions, integrins function in cell  adhesion and migration have been well characterized and are introduced in the following sections.  35  Protein Integrin αPS1 Integrin αPS2 Integrin αPS3 Integrin αPS4-5 Integrin βPS Integrin βν Talin ILK Laminin α3,5 (LanA) Laminin α1,2 (Wb) Laminin β Laminin γ Perlecan Collagen IV  Gene multiple edetamous wings (mew) inflated (if) scab (scb)or volado (vol) αPS4-5 myospheroid (mys) βν integrin (βInt-ν) rhea integrin-linked kinase Laminin A (LanA) wing blister (wb) Laminin B1 (LanB1) Laminin B2 (LanB2) terribly reduced optic lobes (trol) viking (vkg)  Vertebrate homologs Integrin α3, α6 and α7 Integrin αIIB, α5, α8 and αV none none Integrin β1, β2 and β7 None Talin-1 and Talin-2 ILK Laminin α3 and α5 Laminin α1 and α2 Laminin β Laminin γ Perlecan Collagen IV  Table 1: Drosophila integrins. focal adhesion complex components, ECM ligands and their vertebrate homologs.  36 Drosophila integrins in cell adhesion Stable cell adhesion is crucial for maintaining tissue architectures, and is formed directly by cell-cell junctions and indirectly by cell-ECM adhesions.  In Drosophila,  integrins recognize different ECM ligands through the alpha subunits with αPS1βPS for laminins and αPS2βPS for RGD-containing proteins [201, 214].  These two integrins are  frequently expressed by apposed tissues separated by the ECM and mediate indirect cell-cell adhesion [214, 215].  For example, in the Drosophila wing, the dorsal and  ventral epithelial layers are firmly attached to each another through a thin basement membrane (ECM) that forms between them.  The αPS1 and αPS2 subunits are  respectively expressed in the dorsal and ventral layers [198, 216].  Mutants of either α or  β subunits give wing blisters where the epithelial layers separate from each other [216-218]. Another example of integrin mediating cell adhesion is the myotendinous junction (MTJ).  In embryos and larvae, longitudinal muscles attached to each other and to the  animal body wall through the specialized epidermal tendon cells, which form the MTJ and transduce muscle contractions to the cuticle [219].  The extracellular matrix is  deposited between muscles and tendon cells and mediates their connections through cell surface integrin receptors.  αPS1βPS integrin is expressed by the tendon cell, while  αPS2βPS integrin is concentrated at the ends of each muscle [205, 220]. In integrin mutant embryos, somatic muscles are contracted into spheres and are largely detached 37  from the epidermis [205, 221-223].  Studies in the wing and muscle attachment suggest  that different integrins can attach two cell layers to opposite sides of a basement membrane. Another important aspect of integrin mediate adhesions is that integrin complexes are more dynamic than previously thought.  Using "fluorescent  recovery after  photobleaching" (FRAP), the integrin adhesion complexes are shown to undergo turnover in embryonic and larval MTJs [224].  The proportion of integrin complex components  that undergo turnover decreases as the animal ages and accompanying the growth of the MTJ.  This suggests that integrin turnover in cell-ECM adhesion is regulated during  development and is adapted to normal tissue growth and maintenance. Integrin-dependent cell-ECM adhesion is also important for initiation and maintenance of sarcomere integrity in developing and adult muscles [225].  For example,  down-regulation of integrin mediated adhesion in adult fly muscles results in progressive loss of the sarcomere cytoarchitecture and muscle function [226], suggesting that integrin dependent adhesions have roles in more than passive connections. The Drosophila glial cells, especially surface glia are tightly attached to each other and to the ECM layer neural lamella.  However, it has not been shown whether glial  attachment is also dependent on integrins.  In Chapter 2, I show that αPS2βPS and  αPS3βPS are present in different glial layers in the larval peripheral nerve and are required for the formation of different forms of glial ensheathment. 38 Drosophila integrins in cell migration Integrin binding to ECM components mediates not only stable cell attachment to the ECM, but also regulates cell migration on the ECM.  The involvement of Drosophila  integrins in cell migration is implicated from integrin distribution and genetic studies of integrin mutants.  Characterization of integrin dependent cell migration has been  focused on several embryonic events including midgut migration, tracheal system formation, salivary gland development and dorsal closure. Most of the integrin dependent embryonic migration events share a common characteristic in that the ectodermal or endodermal migrating cells move along an mesodermal substratum and the ECM is usually deposited between the migrating cells and mesodermal cells.  For example, the Drosophila midgut is formed by two groups of  endodermal primordial cells which arise separately from anterior and posterior embryonic regions.  Both primordia migrate along a substratum provided by visceral mesoderm  [227, 228]. The prevalent integrin beta subunit βPS is detected in midgut cells [205]. When both maternal and zygotic βPS are removed from the embryo, midgut primordial cell migration is delayed but can eventually complete [209, 229].  A second integrin beta  subunit βν is more specific to the midgut throughout embryonic to pupal stages [206]. The midgut primordia migration is completely blocked only when both integrin beta subunits are removed [230].  This suggests that two integrin beta subunits might  function redundantly to regulate midgut primordial cell migration. 39  A second example is the Drosophila larval tracheal system, which consists of two major trunks and six primary branches and works to conduct oxygen from the exterior to the internal tissues.  The trachea are derived from clusters of embryonic ectodermal cells,  called tracheal placodes, which migrate in distinct and stereotyped directions to form the tracheal branches [231].  In integrin alpha and beta subunit mutants, migration of the  visceral branch, one of the six primary branches, is specifically impaired [232] and the tracheal truck is occasionally disrupted [203]. Another phenomena shared by these integrin dependent migrating events is that the distributions of integrin alpha subunits are tissue specific.  In both the midgut and  trachea, migrating primordial cells express αPS1 and αPS3 and visceral mesoderm expresses only αPS2 [209, 232].  This suggests that different integrins might have  distinct roles in leading cell migration and maintaining the substrate. Glia are highly motile cells in the nervous system and glia can migrate over long distances in the embryonic peripheral nerve [92], the larval eye imaginal disc [124] and the pupal wing [233].  However, it has not been reported that integrins have any role in  these Drosophila glial migrations.  In Chapter 3, I show that integrins are necessary for  glial migration in the larval optic stalk.  1.3.3  Integrin ligands and binding partners  Integrin heterodimers are capable of binding a large number of extracellular ligands 40  and intracellular partners [138, 200]. tissue specific.  Some of these integrin interacting proteins are  For example, in vertebrates cell surface ICAM and VCAM are specific  ligands for leucocyte integrins [200].  Some interacting proteins are organism specific.  For example, an ECM protein Tiggrin necessary for MTJ formation is not observed in organisms other than Drosophila [234].  Some integrin related proteins that are found in  both insects and vertebrates display quite different functions.  An example is focal  adhesion kinase (FAK) that functions downstream of integrins and is an essential protein in mice [235], but dispensable in Drosophila [236].  In the following sections I focus on  the most common integrin ligands and binding partners shared by both invertebrates and vertebrates (Table 1). Laminins Laminins are a family of heterotrimeric glycoproteins assembled by an α-, β- and γsubunits, each encoded by a different gene.  Currently five alpha, three beta and three  gamma subunits are known in mice and humans, which form at least 16 different laminin trimers [237].  Laminins are conserved in most multicellular metazoans.  Four laminin  subunits have been characterized in both C. elegans and Drosophila, which include two alpha, one beta and one gamma subunits [238].  In Drosophila, two laminins are formed  by two different alpha chains. One alpha subunit α3,5 is encoded by Laminin A (LanA) gene [239] and another α1,2 is encoded by wing blister (wb) gene [240]. 41  The alpha  subunits α3,5 and α1,2 are named for their similarity to vertebrate α3 and α5 or α1 and α2 subunits [240].  Both laminins contain a common β1 subunit encoding by Laminin B1  (LanB1) gene [241], and a γ1 subunit producing by Laminin B2 (LanB2) gene [242]. Vetebrate laminins are recognized by at least seven integrins even though only four of them (α3β1, α6β1, α6β4 and α7β1) are cited as laminin-receptors [200].  In  Drosophila, the α3,5 subunit (LanA) containing laminin is bound to PS1 integrin but not PS2 [202].  The α1,2 subunit (wb) contains a RGD sequence which is a potential  binding site for PS2 integrin [243]. Laminins are the major components of the ECM and are broadly distributed in both developing and adult animals.  They are required for the morphogenesis and function of  many organs including the brain, kidney, lung, digits, skin and muscles [238].  In  Drosophila, laminins are widely detected in embryos and larvae [239, 240, 242, 244]. Null mutants of laminin subunits cause embryonic lethality and display morphogenetic defects in heart formation, germ-band extension and retraction, somatic muscle and wing epithelial attachment, gut and trachea formation [240, 245, 246].  These phenotypes are  also seen in integrin mutants [205, 215, 216], supporting that laminins are integrin ligands. Talin Talin is a large polypeptide of more than 2000 amino acids and usually forms an 42  antiparallel homodimer [247]. Talin protein has an elongated shape with an N-terminal globular head containing a FERM domain, and a C-terminal rod formed by 63 helices [248].  Vertebrate genomes host two talin genes talin-1 and talin-2, which encode  similar proteins [249], while one talin gene rhea is found in Drosophila [250]. Talin is an actin binding protein and acts as a physical linker to connect the integrin β subunit to the actin cytoskeleton [251]. The linking function is accomplished through two integrin binding sites (one in the FERM domain and one in the rod) and two actin binding sites (one in the middle of the rod and one at the C-terminal tail) in the talin protein [252]. Talin can also link integrins indirectly to the actin by binding to another actin interacting protein Vinculin [253]. Gene disruption studies in both Drosophila and mice confirm Talin’s essential roles in many integrin-mediated developmental events. Disruption of the mouse talin-1 gene results in an embryonic lethality during gastrulation [254].  Conditional knockout of  talin-1 provides more evidence that Talin-1 is required for integrin mediated platelet function [255, 256] and myotendinous junction integrity [257].  However mice  homozygous for a talin-2 allele that produces truncated Talin-2 protein were normal and healthy [258].  However Talin function is probably not fully displayed unless Talin-1  and Talin-2 are both removed because Talin-2 has been found to compensate the loss of Talin-1 in cultured cells [248].  Then studies on the sole Drosophila Talin might provide  a better understanding of Talin function.  Ablation of Talin in the Drosophila embryo or 43  the wing imaginal disc disrupts the formation of focal adhesions and gives similar phenotypes as the integrin mutants [250, 259].  More sophisticated genetic studies  suggest that Talin plays an essential role, as least in Drosophila, in activating integrins and strengthening integrin-ECM interactions [260].  Surprisingly, genetic studies in  Drosophila also revealed an integrin independent function of Talin in suppressing DE-cadherin expression [261]. Integrin-linked Kinase (ILK) Integrin-linked kinase (ILK) was originally identified as a novel kinase that binds the cytoplasmic tail of β1 integrin [262].  ILK contains four N-terminal ankyrin repeats, an  intermediate PH (pleckstrin homology) domain and a C-terminal kinase domain.  ILK  forms a heterotrimeric complex, called the IPP complex, through its N-terminal domain with the adaptor protein PINCH (Particularly Interesting Cys-His-rich Protein) and through its kinase domain with a family of F-actin binding proteins, parvins [139].  The  IPP complex can interact with other actin binding proteins, such as paxillin, and thus regulate actin cytoskeleton in multiple ways [263]. Genetic studies suggest that ILK is an important mediator downstream of integrins in both mammals and Drosophila. different species.  However the degree of importance is different in  Deletion of ILK in mice causes lethality at embryonic day 5.5-6.5  [264], when β1 integrin homozygous null mutants also die [265, 266]. 44  Conditional  ablation of ILK and integrins generates similar phenotypes in specific tissues.  For  example, loss of ILK or β1 integrin in Schwann cells results in the demyelination in the mouse peripheral nerve [193, 194].  In Drosophila, ILK mutants display only a subset of  phenotypes that are observed in integrin mutants, which are mostly defects at muscle attachment and wing adhesion [267].  In addition, the muscle detachment defects in ILK  null embryos are observed at and post stage 17, which is much later than the more severe detachment phenotypes seen in integrin or Talin mutant embryos [250].  These studies  suggest that ILK has limited functions in maintaining strong adhesions in Drosophila.  45  1.4 Rationale, hypothesis, and objectives Although much evidence indicates that integrin-mediated ECM interactions have important roles in regulating vertebrate glial development, many fundamental questions remain unclear. neurons?  Do glial integrins mediate direct interactions with other glia and  Are there non-ECM ligands for glial integrins? Do glial integrins synergize  with other cell surface receptors, such as: fibroblast growth factors (FGF)? integrins important for non-myelinating Schwann cells?  Are  Answering these questions will  greatly benefit our understanding of how glia and neurons development are orchestrated. However, two difficulties significantly limit further investigations of integrin function in vertebrate glia.  First, manipulating tissue specific disruption of vertebrate genes is  relatively difficult and generation of transgenic animals and mutations are time consuming and costly.  Second, vertebrate integrins are a large family including multiple  alpha and beta subunits, which display complex spacial and temporal expression patterns in glia.  Potential abundance and compensation between different alpha or beta subunits  further complicates the functional examination of a specific integrin.  Investigation of  new genetic models seems necessary to overcome these problems. Drosophila melanogaster is a good alternative to study integrin function in glial development.  First, Drosophila glia share many structural and functional similarities  with vertebrate glia.  Second, a much smaller integrin family has been studied in many  tissues except glia in the fly and show many conserved functions. 46  Thus my hypothesis  was that integrins might also been used by Drosophila glia to regulate different aspects of glial development.  I examined distribution of integrin alpha and beta subunits in  different glial subtypes and conducted loss of function experiments using RNA interference (RNAi) and MARCM (Mosaic Analysis with a Repressible Cell Marker). My major findings are: 1. In Chapter 2 and 3, I show that αPS2βPS and αPS3βPS integrins are expressed by different glial layers in both larval peripheral nerve and optic stalk, where they form adhesion complexes with Talin and ILK. 2. In Chapter 2, I show that integrins necessary for the PG and WG to form proper sheaths in the larval peripheral nerve. 3. In Chapter 3, I show that integrins are required cell autonomously for glial proliferation, migration and barrier formation, and non-cell autonomously for photoreceptor axon outgrowth.  47  Chapter 2  Integrins are necessary for glial ensheathment  formation in Drosophila peripheral nerve1  2.1 Introduction The development of the glial sheath in peripheral nerves is vital to provide structural support, insulate and protect axons from physical damage and pathogens.  The  architecture of peripheral nerves consists of multiple glial layers and a surrounding basal lamina.  Centermost in vertebrate nerves are the myelinating or non-myelinating  Schwann cells that wrap individual or bundles of axons respectively.  Groups of  myelinated or non-myelinated nerve fibers are initially encased by a basal lamina, later becoming embedded in a collagenous connective tissue, the endoneurium, to form fascicles.  Each fascicle is sheathed by the perineurium, a protective barrier of  overlapping squamous-like perineurial cells [268].  Similarly, the Drosophila larval  peripheral nerve is made up of several distinct layers [109].  The innermost wrapping  glia (WG) separate and ensheath axons in a manner similar to vertebrate non-myelinating SCs.  1  The WG are next surrounded by subperineurial glia (SPG) that form intercellular  This chapter has been adapted from one published article: X.Xie, and V. J. Auld. Integrins are necessary for the  development and maintenance of the glial layers in the Drosophila peripheral nerve. Development. Sep 2; 138, 3813-3822, 2011. Article content reproduced here with permission from The Company of Biologists. Copyright 2011 48  septate junctions and create the blood-nerve barrier [109].  The outermost cell layer is a  monolayer of squamous-like perineurial glia (PG) [269].  The final layer is the neural  lamella (NL), a dense basal lamina encasing each peripheral nerve.  Drosophila and  vertebrate peripheral glia express many of the same proteins such as NCAM and L1 [83], homologues of the paranodal junction proteins such as paranodin, contactin and neurofascin [270] and use many of the same developmental programs such as Erb/neuregulin signaling [63, 271]. In vertebrates, the basal lamina contains ECM components known to be important for glial development.  For instance, laminins, a major ECM component are essential for  Schwann cell differentiation, axon sorting and myelination [272-275].  Similarly  non-myelinating Schwann cells lacking laminins fail to differentiate and the associated C-fibers are lost [197].  These ECM signals are transduced by specific receptors  including  Glypican  Dystroglycan,  and  Integrins  [276].  Integrins  are  the  best-characterized ECM- receptors in Schwann cells and consist of one alpha and one beta subunit.  Loss of beta1 integrin or integrin-linked kinase (ILK) in myelinating  Schwann cells results in defects in radial sorting and myelination [193, 194, 277]. While there is evidence for a role of integrin adhesion and signaling in myelinating Schwann cells, the role of integrins in the development of the non-myelinating Schwann cells or the perineurium is not understood. Complicating the investigation of integrins in peripheral nerve development is the 49  complexity of integrin heterodimer expression in vertebrates [276].  In contrast,  Drosophila melanogaster has a relatively simple family of integrin subunits consisting of five alpha subunit and two beta subunits [278].  Therefore I used Drosophila to  investigate the role of integrins and ECM interactions during the development of the glial layers of the peripheral nerve.  In this chapter I show that specific integrin heterodimers  play a role in the development and maintenance of the glial layers.  Down-regulation of  integrin expression results in wrapping defects in both the PG and WG.  Furthermore,  the basal lamina is essential for the proper maintenance of the perineurial wrap. Collectively my results demonstrate that Drosophila integrins and the basal lamina are essential for proper glial sheath development and maintenance in the peripheral nerve.  50  2.2 Methods and materials 2.2.1  Fly strains and genetics  The following fly strains were used in this study: repo-GAL4 [93]; SPG-GAL4 [279]; Nrv2-GAL4 [280]; Gli-GAL4 [281]; 46F-GAL4 (a gift from Dr. Yong Rao); UAS-mCD8::GFP [282]; UAS-mCD8::RFP (a gift from Dr. Elizabeth Gavis); UAS-Mmp2  [283];  UAS-Dicer2  [284];  tubP-GAL80ts  [285];  mys1  [286];  FRT19A,tubP-Gal80,hsFLP,w* [282]; repo-FLP [109]; UAS-nls::GFP (a gift from Dr. Douglas Allan).  The following GFP protein-trap insertions were used:  jupiter::GFP; nrv2::GFP; ILK::GFP; talin::GFP [287, 288].  perlecan::GFP;  UAS-RNAi strains were  obtained from the VDRC (Austria) [284], NIG (Japan), and TRiP (Harvard).  RNAi  experiments were carried out at 25°C and with UAS-Dicer2 in both control and experimental crosses unless specified.  Control and βPS MARCM clones were  generated using: female FRT19A; +/+; repo-GAL4, UAS-CD8::GFP/TM6B, Tb or mys1, FRT19A/FM7; +/+; repo-GAL4, UAS-CD8::GFP/TM6B, Tb with male FRT19A, tubP-Gal80, hsFLP, w*/Y; +/+; repo-FLP.  2.2.2  Immunohistochemistry  The following primary antibodies were obtained from the Developmental Studies Hybridoma bank (DSHB, University of Iowa): mouse anti-βPS (CF.6G11) [198] at 1:10, 51  mouse anti-αPS2 (CF.2C7) [198] at 1:5, mouse anti-Repo (8D12) [289] at 1:50.  Other  primary antibodies were: rabbit anti-αPS3 [290] at 1:200, rabbit anti-HRP (Jackson ImmunoResearch) at 1:500, rabbit anti-Drosophila γ-laminin (LamB2) (Abcam) at 1:100. All secondary antibodies were used at 1:200: goat anti–mouse Alexa568 and Alexa647, goat anti–rabbit Alexa568 and Alexa647 (Molecular Probes/Invitrogen).  Dissection and  fixation for immunofluorescence was performed according to standard procedures [92].  2.2.3  Imaging analysis  Unless specified, images were obtained with a DeltaVision Spectris (Applied Precision), using a 60X objective (NA 1.4), with 0.2 micron Z sections.  Image stacks  were deconvolved and rotated with SoftWorx (Applied Precision), based on measured point spread functions of 0.2 micron fluorescent beads (Molecular Probes) mounted in Vectashield (Vector Laboratories). CS4 for compilation.  Images were exported to Photoshop and Illustrator  For lower magnification images, images were captured with using  a 20X objective (NA 0.4) (Deltavision) or a 10X objective (NA 0.3) (Axioskop 2, Zeiss). PG numbers were counted in nerves from abdominal segment 8 (A8) from late 3rd instar larvae. The mean and the standard deviation were calculated in Excel 2010 (Microsoft).  52  2.3 Results 2.3.1  Glial layers in the Drosophila larval peripheral nerve  In the larval peripheral nerve, axons are wrapped by three consecutive layers of glia: the outermost perineurial glia (PG), intermediate subperineurial glia (SPG) and inner wrapping glia (WG) (Fig. 8A) [109].  The dense basal lamina (neural lamella, NL) that  surrounds the nerve is visualized with the proteoglycan Perlecan endogenously tagged with GFP (Fig. 8B). The PG are located just below the neural lamella and specifically express an enhancer trap line 46F-GAL4 and Jupiter::GFP, a microtubule-associated protein [291] endogenously tagged with GFP (Fig. 8C-D) (Table 2).  Directly below the  perineurial layer are the SPG, which express either SPG-GAL4 [109] or Gliotactin-GAL4 (Gli-GAL4) (Fig. 8D) (Table 2).  In the center of the peripheral nerve are the WG that  ensheath individual axons and axonal bundles.  WG express Nervana2-GAL4  (Nrv2-GAL4) (Fig. 8E) and the Nervana 2 protein tagged with GFP (Nrv2::GFP) (Fig. 8F) (Table 2).  53  Figure 8: Axons are surrounded by three glial layers in the larval peripheral nerve. A) Cartoon showing a cross section of a 3rd instar larval peripheral nerve.  From  external to internal are the neural lamella (NL, black), perineurial glia (PG, light blue), subperineurial glia (SPG, purple), wrapping glia (WG, dark purple) and axons (AX, grey). Two PG nuclei are shown in green. B-F) Orthogonal sections from 3rd instar nerves illustrate the GFP tagged proteins and GAL4 drivers used to label the ECM and different cellular glial layers. digitally expanded.  Panels were  The NL was labeled using Perlecan::GFP (B, green).  PG were  labeled using 46F>CD8::RFP (B and C, red) and Jupiter::GFP (C and D, green).  SPG  were labeled using Gli>CD8::RFP (D and F, red).  WG were labeled using  Nrv2>CD8::GFP (E, green) and Nrv2::GFP (F, green).  Axons were immunolabeled  using an anti-HRP antibody (E, red; F, blue).  All glial nuclei were shown using DAPI  labeling (blue) except in C where the SPG nucleus was immunolabeled using a Repo antibody (blue).  Scale bars are 5 μm.  54  Glia subtypes NL PG  SPG  WG  Repo-GAL4  X  X  X  46F-GAL4  X  GAL4 drivers:  SPG-GAL4  X  Gli-GAL4  X  Nvr2-GAL4  X  GFP trap lines: perlecan::GFP  X  viking::GFP  X  jupiter::GFP  X  nervana2::GF P  X  Table 2: Summary of markers for ECM and glial layers in the peripheral nerve. NL: neural lamella; PG: perineurial glia; SPG: subperineurial glia; WG: wrapping glia.  55  2.3.2  Specific integrins are located in the glia cells of the larval peripheral nerve  I began my investigation of the role of integrins in peripheral nerve development by asking which integrins are expressed by the glia.  I was unable to detect integrin  expression in embryonic glia either by genetic or immunofluorescence analysis (data not shown).  Thus my analysis concentrated on larval peripheral glia and on those integrin  subunits with clear vertebrate homologues (βPS, αPS1, αPS2, αPS3) (Brown et al., 2000). All three classes (PG, SPG and WG) expressed integrin complexes when assayed with an antibody to the beta subunit, βPS (Mys) (Fig. 9).  In addition I observed  immunolabeling with antibodies to two alpha subunits, αPS2 and αPS3 (Fig. 9) but not αPS1 (data not shown).  The integrin subunits were observed in discrete puncta or  stripes and were associated with ILK::GFP (Fig. 9) and Talin::GFP (Fig. 10).  I observed  that the GFP fusion proteins only partially overlapped with βPS integrin (Fig. 9B-D). This was not unexpected as ILK and Talin form adhesion complexes with integrins through binding to the intracellular domain of beta integrin [292] and the integrin antibodies are thought to bind the extracellular domain of these transmembrane receptors [198]. The co-localization of the integrin subunits with ILK and Talin in the larval nerve suggests that adhesion complexes are found throughout the glial layers of the peripheral nerve.  56  57  Figure 9: Integrins and Integrin-Linked-Kinase (ILK) are expressed in the larval peripheral nerve. Antibodies for different integrin subunits were used to immunolabel 3rd instar nerves expressing ILK::GFP.  A and E, single 0.2 μm Z-sections show βPS (A, red), αPS2 (E,  red), αPS3 (E, blue), ILK::GFP (green) and DAPI (A, blue) labeling.  The regions in the  red boxes in panels A and E were digitally expanded and are shown in B-C and F respectively.  The green lines in panels A and E show the positions of orthogonal  sections, which are shown in D and G respectively. Scale bars are 10 μm in A and E, and 5 μm in all other panels. A-D) ILK::GFP (green) and βPS (red) immunolabeling overlap in the peripheral nerve (arrows). Arrowheads point to ILK::GFP that is associated with βPS labeling in the different focal planes shown in B (z=31) and C (z=34). E-G) ILK::GFP (green) is associated with αPS2 (red, arrowheads), αPS3 (blue, arrows).  58  Figure 10: Talin forms adhesion complexes with βPS integrin in the larval peripheral nerve. Antibodies for βPS (red) and HRP (blue) were used to immunolabel 3 rd instar nerves expressing Talin::GFP (green).  Panel A is a single 0.2 μm Z-sections.  the red boxes was digitally expanded and is shown in B. positions of orthogonal sections, which are shown in C. 5 μm in B and C.  59  The regions in  The green line shows the  Scale bars are 10 μm in A, and  2.3.3  The beta integrin subunit is expressed in all glial cell layers  To address which glial layers express different integrin complexes, I used a combination of proteins endogenously tagged with GFP and GAL4 drivers to label the individual glial layers (Fig. 8) (Table 2).  First, I examined the distribution of the  common βPS subunit, which is able to form heterodimers with all alpha subunits.  PG  were labeled with Jupiter endogenously tagged with GFP fusion in conjunction with SPG labeled using Gli-GAL4 driving the expression of UAS-CD8::RFP (Gli>CD8::RFP). Immunolabeling of βPS integrin was observed on the PG outer membrane (Fig. 11C-D) suggesting an interaction of the integrin complex with the external ECM/neural lamella [109].  βPS integrin immunolabeling was also found between Jupiter::GFP and  Gli>CD8::RFP (Fig. 11C-D), suggesting that these integrin proteins also localize to the PG and SPG boundary. The βPS was also observed in the nerve, internal to the CD8::RFP labeled SPG (Fig. 11C-D) in the areas occupied by WG and axons.  To study whether the internal βPS  integrin is located to the WG membrane, Nervana2 endogenously tagged with GFP (Nrv2::GFP) [288] was used to label the WG (Fig. 11E-G). Gli>CD8::RFP was used to label the SPG membrane.  At the same time  βPS immunolabeling was  consistently found to associate with both the Gli>CD8::RFP labeled SPG (Fig. 11F-G, arrowheads) and with Nrv2::GFP.  βPS was often observed between Nrv2::GFP and  protrusions of CD8::RFP (Fig. 11F-G) and thus might mediate interactions between the 60  SPG and WG. Importantly, all of the βPS integrin labeling observed in the peripheral nerve appears to be expressed by glial cells but not neurons.  I confirmed this observation by  knock-down of βPS expression using βPS-RNAi specifically in the glia cells (see following sections).  My data indicate that βPS integrin is expressed by the glial cells  and forms adhesion complexes in the glial membranes in the larval peripheral nerve.  61  Figure 11: βPS integrin is expressed in the peripheral glia layers. Immunolabeling for the βPS integrin subunit in 3rd instar nerves expressing different glial markers in single 0.2 μm Z-sections.  Different glial layers were labeled with specific  markers: PG using Jupiter::GFP (B-D, green), SPG using Gli>CD8::RFP (B-G, red) and WG using Nrv2::GFP (E-G, green).  The labeled glial layers are also illustrated in  cartoons to the right. The red boxes in panels B and E were digitally expanded and shown in panels C and F respectively.  The green lines in panels B and E show the positions of  orthogonal sections shown in D and G respectively. βPS integrin was found in: the PG outer membrane (C and D, arrow), in the PG-SPG boundary (C and D, arrowhead), the SPG inner membrane (F and G, arrowhead), and the WG membrane (F and G, arrows). Scale bars are 10 μm in A and D, 5 μm in B-C and E-F.  62  2.3.4  Differential glial expression of integrin alpha subunits  Both the αPS2 and αPS3 integrin subunits were located to adhesion complexes with ILK::GFP and Talin::GFP in the larval peripheral nerve and are expressed with different intensities in the different glia layers (Fig. 9E-G and data not shown).  The specific glial  layers markers outlined above were used to further characterize the discrete localization of the alpha subunits (Fig. 12).  αPS2 integrin immunolabeling was found on the  external surface of the nerve, outside of the Jupiter::GFP labeled PG similar to βPS (Fig. 12A-C).  This further supports a role for the integrin complex mediating adhesion  between the PG and the ECM.  αPS2 was also found between Jupiter::GFP and  Gli>CD8::RFP (Fig. 12A-C), indicating that αPS2 is also localized at the PG and SPG boundary.  A few αPS2 positive puncta were also found in the center of the nerve,  internal to the Gli>CD8::RFP (Fig. 12A-C), suggesting that αPS2 might be expressed at low levels by the internal WG or extensions of the SPG into the center of the nerve. Conversely most of the αPS3 integrin immunolabeling was found in the center of the nerve internal to the Gli>CD8::RFP (Fig. 12D-F).  αPS3 was located between the RFP  labeled SPG processes and Nrv2::GFP labeled processes.  Moreover, αPS3 was  associated with CD8::RFP labeled membranes seen protruding between the internal Nrv2::GFP labeled members of the PG (Fig. 12D-F), suggesting that αPS3 might be involved in a SPG and WG interaction.  However I cannot rule out that αPS3 expression  may indicate an interaction between the WG and their associated axons. 63  In general, my  data suggest that the αPS2 and αPS3 integrin subunits are both expressed by the different glial layers and might be important for glia-ECM and glia-glia interactions.  64  65  Figure 12: αPS2 and αPS3 integrins are expressed in different glial layers of the larval peripheral nerve. The 3rd instar larval nerves were immunolabeled with antibodies for αPS2 (A-C, blue) and αPS3 (D-F, blue) integrins in single 0.2 μm Z-sections.  Specific glial layers were  labeled with different markers: PG using Jupiter::GFP (A-C, green), SPG using Gli>CD8::RFP (A-F, red) and WG using Nrv2::GFP (D-F, green). The labeled glial layers are also illustrated in cartoons to the right. magnified and shown in B and E.  The red boxes in A and D were digitally  D and G are orthogonal sections of Z-stack from the  positions shown with green lines in B and E. Scale bars are 10 μm in A and D, 5 μm in B-C and E-F. A-C) αPS2 integrin labeling is mostly found in the outer glial cells, in the PG outer membrane (arrows), in the PG-SPG boundary (arrowheads), and in the SPG inner membrane (asterisks). D-F) αPS3 integrin labeling is mostly found in the internal glial layers, in the SPG inner membrane (arrowheads) and in the WG membrane (arrows).  αPS3 integrin  labeling was often associated with the SPG membrane protrusions (asterisks).  66  2.3.5  Integrin function is necessary in larval peripheral glia  The question then arises, what functions do integrins have in the peripheral glia? Null mutants of βPS (mys) are embryonic lethal and I was unable to detect βPS immunolabeling in embryonic glia (data not shown).  In mys mutants, I was unable to  detect any defects in glial cell migration or defasciculation of peripheral nerves (a hallmark of disruption in glial ensheathment) (data not shown).  Therefore I took two  approaches to remove βPS function from peripheral glia during larval development, MARCM analysis and RNA interference (RNAi). MARCM analysis in the glia To generate mosaic clones for a βPS (mys) null in peripheral glia I used the MARCM technique and assayed glial development in 3rd instar larvae.  CD8::GFP labeled mys  clones were seen in the peripheral (Fig. 13A) and central nervous systems (Fig. 13A). Using this approach the majority of glial clones were in the PG (14/14 in control larvae and 14/16 in mutant larvae).  In control, wild-type clones the CD8::GFP labeled PG  formed a thin layer on the outside of the HRP-labeled axons (Fig. 13B) resulting in an In controls, βPS immunolabeling was  intact circle around the nerve (Fig. 13D).  consistently found in the perineurial membrane (Fig. 13C-D) and internal glial layers. In the null βPS (mys) clones, βPS labeling was absent from the GFP labeled perineurial 67  membrane (Fig. 13E-G) but still present in the internal wild type SPG and WG layers. Loss of βPS integrin resulted in PG that failed to form a uniform sheath such that CD8::GFP labeled membranes were observed only on one side of the nerve (Fig. 13E), and failed to encircle the nerve (Fig. 13G).  It is important to note that the CD8::GFP  labeled membranes of the βPS mutant PG still covered a significant distance along the length of some nerves, suggesting that without βPS integrin, PG were able to send out processes along the nerve.  While MARCM analysis revealed the importance of βPS  integrin in larval PG development, this approach had two disadvantages. majority of glial clones were in the PG.  First, the  SPG (n=0) and WG (n=2) clones were rare, as  these glial subtypes do not actively divide at later stages.  Second, in each nerve with  PG clones, only a small population of PG were βPS mutants, making it difficult to evaluate the overall impact.  68  69  Figure 13: MARCM analysis of βPS mutant in the larval peripheral nerve. MARCM using flippase expression in glia was used to generate wild type or homozygous βPS (mys) clones labeled with CD8::GFP. A) Control and βPS glial clones were identified with CD8::GFP and immunolabeled for Repo (red) and HRP (blue).  At low magnification (10X objective), clones were seen  in both peripheral nerves (arrows) and the CNS (arrowheads). B-G) High magnification images of nerves having control and mutant PG clones are shown in single 0.2 μm Z-sections (B and E) and orthogonal sections of Z-stacks (D and G).  Green lines point to the positions of orthogonal sections.  Regions in the red boxes  were digitally expanded and shown in C and F. B-D: In control clones, CD8::GFP labeling wraps around the circumference of the nerve.  Arrows indicate the βPS labeling (red) in the PG outer membrane compared to  the core of HRP-labeled axons (blue). E-G: In βPS (mys) clones, CD8::GFP labeling is found only on one side of the HRP labeling in the Z-section.  The CD8::GFP labeled membrane does not wrap around the  nerve in the orthogonal section.  βPS labeling is not found in the PG outer membrane  (arrowheads) but is still seen in the inner layers of glia within the nerve. Scale bars are 10 μm in B and E, 5 μm in C-D and F-G.  70 RNAi analysis in the glia To overcome these limitations, I used RNA interference (RNAi) [293] to knock down integrins in all or individual glial subtypes.  RNAi lines known to target specific  integrin subunits [226] were obtained that target at least two independent regions of each integrin subunit.  I tested the ability to knock down specific integrin subunit in glia by  using the repo-GAL4 driver.  Repo is a transcriptional factor exclusively expressed in all  glia except the midline glia [294].  When repo-GAL4 was used to drive βPS-RNAi, the  amount of immunolabeling detected using the βPS antibody was decreased (Fig. 14B) compared to control nerves (Fig. 14A).  The degree of βPS integrin immunolabeling in  the underlying body wall muscles was used as an internal control.  This suggests that  βPS-RNAi is able to knock down βPS expression specifically in glia.  In addition the  loss of βPS labeling throughout the nerve in repo>βPS-RNAi larvae suggests that βPS integrin is expressed only by glial cells and not the neurons. αPS2-RNAi and αPS3-RNAi lines were used to knock down expression of the corresponding alpha subunits.  In repo>αPS2-RNAi nerves, αPS2 labeling was removed  but the αPS3 labeling appeared normal (Fig. 14D). Consistently, αPS3 labeling but not αPS2 labeling was greatly reduced in repo>αPS3-RNAi nerves (Fig. 14E), indicating that the αPS2-RNAi and αPS3-RNAi constructs were subunit-specific.  Labeling of muscles  and trachea were used as internal staining controls for αPS2 and αPS3 respectively. The knock-down of the integrin complex in glial cells caused glial defects and 71  decreased viability. pupal stages.  For instance, repo>βPS-RNAi animals died in late 3rd instar and  When UAS-CD8::GFP was co-expressed to label the glial membrane, I  observed morphological changes paired with reduced intensity and changes in the distribution of CD8::GFP (Fig. 14B). glial cells in the larval peripheral nerve.  This suggests that βPS integrin is necessary in However, repo-GAL4 is expressed in all three  glial layers, which are functionally and morphologically distinct.  Therefore to test for  integrin function in the individual glial layers, different GAL4 drivers were used to express integrin RNAi in specific glial subtypes.  72  73  Figure 14: Expression of integrin-RNAi in glial cells of the peripheral nerve. repo-GAL4 was used to express UAS-CD8::GFP (green) and UAS-integrin-RNAi with immunolabeling for specific integrin subunits in 3 rd larval peripheral nerves. are projections of the entire Z stack.  All panels  Scale bars are 5 μm.  A-B) βPS (red) immunolabeling was observed in the peripheral nerve (N) and muscle from control (A) and repo>βPS-RNAi (B) larvae. HRP (blue).  Axons were labeled with  βPS labeling (red) was greatly decreased in the nerve but unchanged in  the muscle (M). The CD8::GFP (green) and HRP labeling were not evenly distributed along the nerve compared with control nerves (A and B). C-D) αPS2 (red) and αPS3 (blue) immunolabeling was observed in the peripheral nerve, muscle and tracheal (T) from control (C), repo>αPS2-RNAi (D) and repo>αPS3-RNAi (E) larvae.  αPS2 but not αPS3 labeling was decreased in the  repo>αPS2-RNAi peripheral nerve and αPS2 labeling was still found in a muscle.  αPS3  but not αPS2 labeling was decreased in the repo>αPS3-RNAi peripheral nerve and αPS3 labeling was still observed in a tracheal.  74  2.3.6  Integrin function is necessary for perineurial glia wrapping  To study how loss of integrin affects the outer PG layer, 46F-GAL4 was used to express the integrin-RNAi constructs.  In control 3rd instar larvae, 46F-GAL4 driven  CD8::GFP formed a seamless sheath around the exterior of the peripheral nerves (Fig. 15A-D).  The coverage of 46F>CD8::GFP on the entire surface of HRP labeling  suggests that the membranes from different perineurial cells attach to each other and wrap In 3rd instar larvae with 46F-GAL4 driving the  around the nerve collaboratively.  expression of βPS-RNAi (46F>βPS-RNAi), the labeling of the βPS antibody was decreased in the PG (Fig. 15F-H) compared with controls (Fig. 15B-D).  The βPS  labeling appeared normal in the internal glia, indicating that βPS integrin expression was knocked down specifically in the PG. βPS-RNAi caused similar morphological changes to the PG as observed in the βPS null MARCM clones.  The CD8::GFP labeled PG membrane was observed only on one  side of the nerve (Fig. 15F) and failed to encircle the nerve (Fig. 15H).  Moreover,  individual PG cells became distinct (Fig. 15E) as the PG became detached from each other. Given that αPS2 integrin labeling was mostly observed on both sides of perineurial cells, I asked whether loss of αPS2 integrin would phenocopy the βPS-RNAi. 46F-GAL4 was used to express UAS-αPS2-RNAi and similar PG wrapping phenotypes were observed (data not shown).  However the penetrance of the phenotypes was much 75  lower in 46F>αPS2-RNAi larvae compared to 46F>βPS-RNAi larvae (Table 3). severity of the defects was also different.  The  In 46F>αPS2-RNAi larvae, all defective  nerves fell into category 1 in which a minority of PG failed to wrap.  While in  46F>βPS-RNAi, most of the defective nerves fell into category 2 in which a majority of perineurial cells displayed a wrapping failure.  Given the efficiency of UAS-αPS2-RNAi  (Fig. 14D), the difference between 46F>αPS2-RNAi and 46F>βPS-RNAi might due to compensation by another alpha subunit.  As the αPS3 integrin was also found in the  peripheral nerve, UAS-αPS3-RNAi was co-expressed with UAS-αPS2-RNAi using 46F-GAL4.  As expected, more nerves (92%) had PG that failed to wrap and most of  them (69%) were scored as category 2 (Table 3).  This suggests that αPS2 and αPS3  integrins are functionally redundant in the PG similar to observations in midgut formation [209]. To determine if PG wrapping requires the integrin adhesion complex, I specifically knocked down Talin using RNAi driven by 46F-GAL4 and observed wrapping defects similar to the integrin RNAi (Table 3, Fig. 15I-K).  My data suggested PG wrapping of  the peripheral nerve requires an integrin adhesion complex that contains αPS2βPS or αPS3βPS integrin plus Talin. To check if the PG wrapping defect is due to an insufficient number of PG to match the growing nerve surface, I quantified the PG number in A8 nerves by coexpressing a nuclear localized GFP (UAS-nls::GFP) using 46F-GAL4. 76  Control nerves had 40 ±6 PG  on average (n=12).  Knock-down of βPS resulted in either a reduction or increase in PG  nuclei depending on the RNAi line used (GD, 13 ± 3, n=11)(R1, 57 ± 7, n=14). However knock-down of Talin also showed an increase in the number of PG (GD, 50 ±7, n=12)(R1, 55 ± 8, n=14).  As all these RNAi lines generate the glial wrapping  phenotype these changes in glial number are likely not responsible for the failure of the PG to wrap the growing nerve.  77  78  Figure 15: Expression of βPS-RNAi and talin-RNAi in the perineurial glia 46F-GAL4 was used to express CD8::GFP (green), βPS-RNAi (E-H) and talin-RNAi (I-K) in the perineurial glia (PG).  Axons were labeled by HRP staining (blue).  A&E) Lower magnification views of control and mutant nerves are shown in A and E, respectively.  Repo labeling (red) shows all glial nuclei.  Notice the seamless coverage  of GFP in control nerves (arrow) and discontinuity of GFP in mutant nerves (arrowheads). Panels A and E were captured with a 20X objective. B-D, F-K) High magnification images of control and RNAi nerves are shown in single 0.2 μm Z-sections (B, F and I) and orthogonal sections of Z-stacks (D, H and K). Green lines point to the positions of orthogonal sections.  Regions in the red boxes were  digitally expanded and shown in C, G and J. In the control nerve (B-D), CD8::GFP labeling wraps around the circumference of the nerve.  Arrows indicate the βPS labeling (red) in the PG outer membrane.  In the  46F>βPS-RNAi (F-H) or 46F>talin-RNAi (I-K) nerves, CD8::GFP labeling was found only on one side of the nerve and failed to wrap the circumference.  βPS labeling was  reduced from the PG membrane in the 46F>βPS-RNAi nerve (arrowheads in G and H) but seemed normal in 46F>talin-RNAi nerve (arrowheads in J and K).  The labeled glial  layers and morphological changes are illustrated in cartoons to the right. Scale bars: A and E - 50 μm, B, F and I - 10 μm, small panels - 5 μm.  79  No. 1 2 3 4 5 6 7 8 9 10  Drosophila lines  Total No. of nerves  w[1118] (control) UAS-βPS-RNAi (VDRC_GD) UAS-βPS-RNAi (NIG) UAS-αPS2-RNAi (VDRC_GD) UAS-αPS2-RNAi (NIG) UAS-αPS3-RNAi (VDRC_GD) UAS-αPS3-RNAi (NIG) UAS-αPS2-RNAi + UAS-αPS3-RNAi UAS-talin-RNAi (VDRC_GD) UAS-talin-RNAi (NIG)  Category 0 No. of nerves Ratio  Category 1 No. of nerves Ratio  Category 2 No. of nerves Ratio  69  67  97%  2  3%  0  0%  64  0  0%  2  3%  62  97%  59  2  3%  12  20%  45  76%  63  53  84%  10  16%  0  0%  53  49  92%  4  8%  0  0%  61  59  97%  2  3%  0  0%  57  54  95%  3  5%  0  0%  39  3  8%  9  23%  27  69%  51  0  0%  16  31%  35  69%  65  0  0%  3  5%  62  95%  Table 3: Integrin and talin RNAi are expressed by 46F-Gal4. Category 0: category of nerves in which no perineurial glia show wrapping defects; Category 1: category of nerves in which minority of perineurial glia show wrapping defects;  Category 2: category of nerves in which majority of perineurial glia show  wrapping defects; VDRC_GD:  GD constructs from Vienna Drosophila RNAi Center;  NIG: Fly Stocks of National Institute of Genetics in Japan.  80  2.3.7  The neural lamella is required by the perineurial glia Most integrins function through binding to the ECM including αPS2βPS integrin  in Drosophila [295]. Integrins are in turn important for ECM deposition, such as in epithelial morphogenesis [296].  However, when βPS integrin was knocked down in all  glial layers using repo-GAL4, Perlecan::GFP and γ-Laminin labeling appeared normal in RNAi nerves as in control nerves (Fig. 16). This suggests that glial integrins are not required for neural lamella (NL) formation. The question is then, whether integrin function in the PG relies on the NL.  The  major components of the NL, such as Collagen IV, Perlecan and laminins, are deposited by migrating hemocytes during embryogenesis [109, 121].  Consistently, expression of  perlecan (trol) and collagen IV (viking) RNAi in glia did not affect or remove the NL (data not shown).  Instead, I choose to remove the entire ECM layer by expressing a  matrix metalloproteinase, Mmp2, in glial cells.  Mmp2 is one of two Mmp family  members in Drosophila [283] and is attached to the extracellular membrane through a glycophosphatidylinositol (GPI) anchor [297].  Constitutive expression of Mmp2 by  repo-GAL4 or 46F-GAL4 caused lethality at embryonic and larval 1 st instar stages. Therefore to study the function of the NL in later stages, a temperature sensitive inhibitor of GAL4 (GAL80ts) was used.  Embryos and larvae were raised at the permissive  temperature (18°C) and at the early 3rd instar stage.  Mmp2 expression was activated by  transferring the larvae to the restrictive temperature (29°C) for one day prior to dissection. 81  The ECM was labeled using Perlecan::GFP (or Collagen IV::GFP, data not shown) and Laminin immunolabeling to determine the effectiveness of the Mmp2.  In control nerves,  Perlecan::GFP and γ-Laminin were both found in the NL surrounding the PG (Fig. 17A-E).  When Mmp2 was expressed by the PG, Perlecan::GFP and Laminin were  greatly reduced and became detached from the PG (Fig. 17F-J) suggesting that over-expression of Mmp2 in PG was sufficient to remove the NL.  Surprisingly, high  levels of Laminin immunolabeling were observed within the PG (Fig. 17I-J) suggesting that glia might up-regulate Laminin to compensate the loss of the NL. The PG (46F>CD8::RFP) expressing Mmp2 did not form a complete circle around the nerve (Fig. 17H, J).  Thus disruption of the ECM by Mmp2 generated a wrapping  defect in the PG similar to that of the knock down of βPS in the 46F>βPS-RNAi nerve. Remarkably the short-term expression of Mmp2 in the PG resulted in a strong CNS phenotype in which the ventral nerve cord became skinny and elongated compared to controls (Fig. 18).  This suggests a role of ECM and PG in maintaining nervous system  morphology by constricting the ventral nerve cord.  82  Figure 16: Expression of βPS-RNAi in glial cells does not change the neural lamella in peripheral nerves. Repo-GAL4 was used to express βPS-RNAi in all three glial layers and NL was labeled with Perlecan::GFP (green) and γ-Laminin immunolabeling (red). 0.2 μm Z-sections of the nerves.  A and D are single  B and F are projections of the entire Z-stack for each.  Regions in the bed boxes were digitally expanded and shown in B and E. βPS labeling was greatly decreased in Repo> βPS-RNAi nerves (blue in D-F) compared with that in control nerves (blue in A-C).  βPS was also found in tracheal (T) and  muscles (M), and it was not changed in RNAi larvae. Perlecan::GFP and γ-Laminin labeling appeared to be same in control nerves and RNAi nerves suggesting no change to NL when βPS integrin was removed. Scale bars are 10 μm in the big panels, and 5 μm in the small panels. 83  84  Figure 17: Overexpression of Mmp2 degrades the neural lamella and causes perineurial glial wrapping defects. 46F-GAL4 was used to express CD8::RFP (red) and Mmp2 in the PG.  The expression  was regulated by a temperature sensitive GAL80 under the control of a ubiquitous tubulin promoter (tubP-GAL80ts). γ-Laminin (blue).  The NL was labeled by Perlecan::GFP (green) and  The labeling and changes to the NL and PG are also shown in  cartoons at the bottom.  A-B and F-G are single 0.2 μm Z-sections of the nerves.  and H are projections of the entire Z-stack. digitally expanded and shown in D and I. from the positions shown with green lines.  C  Areas highlighted by the red boxes were E and J are orthogonal sections of Z-stacks Scale bars: large panels - 10 μm, small  panels - 5 μm. A-E) In a control nerve, Perlecan::GFP (green) and γ-Laminin (blue) labeled the outermost NL (arrows), which surrounds the CD8::RFP labeled PG (red).  Low levels of  Laminin appeared diffuse within the nerve. F-J) In a 46F>Mmp2 nerve, both Perlecan::GFP (green) and Laminin immunolabeling (blue) indicated that the remaining NL is partially removed and detached from the surface of the nerve (arrows).  γ-Laminin immunolabeling appeared increased  within the CD8::RFP (red) labeled PG (arrowheads).  The PG no longer wrapped around  the entire circumference of the nerve and borders of distinct, individual PG were identified.  85  Figure 18: Overexpression of Mmp2 in perineurial glia changes the morphology of the central nervous system (CNS). The perineurial glial coverage of the nervous system was labeled using 46F>CD8::GFP. The length of the CNS increased greatly in larvae expressing Mmp2 in the perineurial glia (46F>Mmp2) (B, arrows) compared with a control larva (A, arrows). PN: peripheral nerve.  86  2.3.8  Integrin function is necessary for wrapping glia ensheathment  In the larval nerve, βPS integrin is not only found in the outer glial cells, but also observed in the central regions where the WG ensheath axons (Fig. 11D-F).  To study  integrin function in the WG, the Nrv2-GAL4 driver was used to knock down βPS in GFP-labeled WG (Fig. 19).  In wild type nerves, the WG (n=72) labeled with CD8::GFP  formed complex processes along the nerve (Fig. 19B).  The CD8::GFP labeling filled in  the space between the HRP label axons (Fig. 19A-C) as each WG produces extensive processes that wrap around axons.  In the Nrv2>βPS-RNAi larvae, CD8::GFP still  formed processes in between HRP labeling and spread along the nerve. complexity of the processes was greatly decreased.  However, the  For instance in the βPS-RNAi line  (VDRC_GD29619), almost 100% WG (n=90) had severe defects in which CD8::GFP labeling was observed in a single longitudinal process (Fig. 19D-F).  Importantly, this  phenotype was successfully repeated in the two βPS null MARCM clones that I was able to obtain in the WG (data not shown).  This suggests that the WG are unable to form or  to maintain complex processes when βPS integrin is reduced. While βPS had a clear function in the WG, I was unable to detect the presence of ECM components in the internal regions of the peripheral nerve.  For example,  Perlecan::GFP (Fig. 17A) and Viking::GFP (Collagen IV) (data not shown) were observed exclusively in the NL with no internal labeling. of Laminins.  I also assayed the distribution  Drosophila has two alpha genes but only one beta (LanB1) and one 87  gamma (LanB2) gene.  Strong beta and gamma laminin expression was detected in the  neural lamella but only diffusely in internal regions of the nerve (Fig. 17B, D, E). However, I cannot rule out the possibility that some subtle ECM might be deposited internally given the diffuse laminin labeling I observed.  Therefore to test if an internal  ECM is necessary for the ensheathment of peripheral axons, Mmp2 was over-expressed in the WG using Nrv2-GAL4.  Unlike the repo-GAL4 or 46F-GAL4, constitutive  expression of Mmp2 with Nrv2-GAL4 did not cause lethality or a decrease in the WG processes in the peripheral nerve (Fig. 19G-H). This suggests that the internal βPS integrin might function in the WG by binding to an unknown ligand that is not an ECM component.  88  89  Figure 19: Expression of βPS-RNAi in the wrapping glia. Nrv2-GAL4 was used to express CD8::GFP (green) and βPS-RNAi in the WG. Anti-HRP and DAPI labeling was used for axons (red) and nuclei (blue), respectively. The labeling and phenotypes observed in the WG are illustrated in the cartoons to the right.  A and D are single 0.2 μm Z-sections; B, E, G and H are projections of the entire  Z-stack.  C and F are orthogonal sections from the positions indicated by the green lines.  A-C) In a control nerve, CD8::GFP surrounded the DAPI labeled glia nucleus (blue) and extended along the nerve to fill complex processes (arrows) in between the HRP labeling (red, arrowheads). D-F) In a Nrv2>βPS-RNAi nerve, CD8::GFP was found only in a single process that extended along the labeled axons (red, arrowheads).  Arrows point to small projections  and puncta of CD8::GFP. G-H): In a Nrv2>Mmp2 nerve, the CD8::GFP labeled processes of the WG appeared normal and spread throughout the labeled axons similar to the control nerve. nuclei were identified with Repo immunolabeling (blue in G). Scale bars: large panels - 10 μm, small panels - 5 μm.  90  Glial  2.4 Discussion 2.4.1  The integrin complex in peripheral glia  Peripheral nerves in vertebrates and Drosophila are organized in similar ways with central axons wrapped by an inner class of glia that are surrounded in turn by layers of external glia.  The glia and the surrounding ECM establish a tubular sheath to protect  axons from physical damage and pathogens [80].  I show that at least two integrin  heterodimers are expressed in these different glial layers and are localized at focal adhesions with Integrin-linked kinase (ILK) and Talin.  I found that αPS2βPS integrin is  prevalent in the PG and the αPS3βPS integrin is more prevalent in the WG.  Since αPS2  integrin is expressed mostly in the outermost PG and can bind ligands containing the tripeptide RGD sequence [234, 286], it most likely functions by binding to ECM ligands in the neural lamella.  The majority of αPS3 integrin is expressed by the internal SPG  and WG and αPS3 integrin has been shown to interact with laminins in Drosophila ([203, 204].  However it is possible that the integrin complex in the WG might have other  ligands and mediate direct cell-cell interactions as I failed to detect a pronounced basal lamina associated with the peripheral axons and Mmp2 expression had no effect in the internal regions of the peripheral nerves.  91  2.4.2  Integrin-ECM interactions are required for PG function  PG form the outermost glial layer in Drosophila nervous system but their origin and function is not well understood even though they have been identified for some time [116, 298].  Drosophila PG are structurally similar to their vertebrate counterparts and have  been proposed to have similar roles [269].  Vertebrate perineurial cells and associated  collagen fibers provide important mechanical support to nerves and their development may rely on ECM mediated signals since β1 integrin is found in the perineurium [193]. However, little is known about the role of integrins in perineurial cells. My results show that Drosophila PG express αPS2 and βPS integrin subunits (and to a lesser extent αPS3), which colocalize with both ILK and Talin.  Knocking down  integrins disrupts perineurial wrapping but I was unable to distinguish whether the PG failed to initiate wrapping or failed to maintain their processes around the nerves to accommodate the growing nerve surface.  However degradation of the neural lamella by  Mmp2 overexpression generates a PG wrapping phenotype similar to βPS RNAi.  Thus  binding of βPS integrin to ligands in the ECM mediates the radial spread of PG around the tubular structure of the nerve and suggests that PG retract their membrane when integrin-ECM interaction is interrupted. produces a similar wrapping defect.  Moreover knock down of Talin in the PG  This suggests that in the PG, integrin adhesion  complexes mediate the connection between the extracellular neural lamella and the intracellular actin cytoskeleton, and are required for the initiation or maintenance of glial 92  ensheathment. The function of the PG in the peripheral nerve is not well understood.  The PG do  not generate an impermeable barrier and it is the SPG layer that creates the blood-nerve barrier [109].  Loss of the PG ensheathment did not result in paralysis or lethality, which  are signs of a disrupted blood-brain barrier [108, 279]. Larger molecules (~ 500 kDa) are blocked by either the neural lamella or the PG [109], perhaps mirroring the protective function of the vertebrate perineurium against pathogens [80].  However, in Mmp2  overexpressing larvae, the affected nerves do become thin and are difficult to keep intact during tissue preparation.  This suggests that neural lamella and PG provide important  mechanical support as is seen with the perineurium in mammalian nerves [80].  2.4.3  βPS integrin is required to maintain wrapping glia ensheathment of the  peripheral axons In the center of Drosophila peripheral nerves, the WG embed axons in bundles or individually within single membrane wraps, similar to the non-myelinating Schwann cells in vertebrate peripheral nerves [299].  My results show that WG predominantly express  αPS3and βPS integrin subunits in complexes positive for ILK and Talin along the WG membrane.  When βPS expression is knocked down, the complexity of the glial  processes between the associated axons is greatly reduced.  Only a few long processes  and small membrane protrusions are observed around the axons, suggesting the WG may 93  be retracting their processes in the absence of βPS integrin.  The role of integrins in WG  appears to be conserved with those seen with Schwann cells in vertebrates.  For instance,  myelinating Schwann cells lacking β1 integrin or ILK do not extend membrane processes around axons resulting in impaired radial sorting [193, 194]. The role of integrins in non-myelinating Schwann cells is not known but my results suggest that integrins have similar functions in Drosophila and vertebrates to mediate glial ensheathment of peripheral axons. No clear ECM has been observed in internal regions of Drosophila nerves either through immunofluorescence or TEM analysis [109, 269].  Therefore, the intregin  complex could promote WG sheath formation by mediating direct cell-cell adhesion between the glial membrane and its associated axon, or between glial membranes.  A  potential candidate for an integrin interacting protein expressed on axons or glia is Neuroglian, the Drosophila L1 homologue.  L1 is an IG domain transmembrane protein  known to bind RGD-dependent integrins [300, 301].  Loss of the integrin binding  domain of L1 results in wrapping defects in both myelinating and non-myelinating Schwann cells [302].  Conversely I cannot rule out the presence of low levels of ECM  given the weak laminin immunolabeling I observed.  Even though Mmp2 expression in  the WG had no effect it is still possible that the integrin mediated adhesion is Mmp2 resistant. To answer this question, further ultrastructural and genetic studies will be necessary. 94  2.4.4  Summary  Peripheral nerve development involves multiple classes of glia that cooperate to form overlapping glial layers paired with the deposition of a surrounding extracellular matrix (ECM).  The formation of this tubular structure protects the ensheathed axons from  physical and pathogenic insult plus changes in the ionic environment.  Integrins, a major  family of ECM-receptors, play a number of roles in the development of myelinating Schwann cells, one class of glia ensheathing the peripheral nerves of vertebrates. However the identity and role of the integrin complexes utilized by the other classes of peripheral nerve glia has not been determined in any animal.  Here I show that in the  peripheral nerves of Drosophila melanogaster, two integrin complexes (αPS2βPS and αPS3βPS) are expressed in the different glial layers and form adhesion complexes with integrin-linked kinase (ILK) and Talin.  Knock down of the common beta-subunit (βPS)  using inducible RNAi in all glial cells results in animal lethality and glial defects. Analysis of integrin complex function in specific glial layers showed that loss of βPS in the outer most layer of glia (the perineurial glia) results in a failure to wrap the nerve, a phenotype similar to MMP2 mediated degradation of the ECM.  Knock down of βPS  integrin in the inner most wrapping glia causes a loss of glial processes around axons. Together my data suggests that integrins are employed in different glial layers to mediate the development and maintenance of the protective glial sheath in Drosophila peripheral nerves. 95  Chapter 3  Integrins are required for glial and neuronal  development in Drosophila visual system  3.1 Introduction Glia provide neurons with metabolic and structural support as well as play a direct role in guiding neuronal development [83, 271, 303].  Many aspects of glial  development are intricately tied to neuronal generated signals through either direct cell-cell contact or secreted factors [63]. Other than neurons, regulatory signals are also provided to glia by the extracellular matrix (ECM) and by neighboring glia.  However,  how glial development and function are modulated by these signals is not completely understood.  A key family of ECM receptors, the integrins, are important for both  neuronal and glial development in the vertebrate nervous system [153, 304]. Integrins are heterodimeric transmembrane proteins, composed of one alpha and one beta subunit that control a wide range of cellular responses including adhesion, migration and differentiation [200].  Ligand binding leads to the recruitment and formation of focal  adhesions, which consist of clustered integrins and focal adhesion proteins including talin, vinculin, tensin, paxillin, and focal adhesion kinase (FAK)[305].  Talin, a key  component of the adhesion complex, binds the cytoplasmic tail of -integrins and functions to link integrins and the actin cytoskeleton [306, 307]. In glial development, integrins are a critical component in oligodendrocytes and 96  Schwann cells during radial sorting of axons and subsequent myelination [169-171, 193, 195, 308, 309].  Similarly loss of FAK or Integrin-linked kinase (ILK) leads to  phenotypes similar to β1 integrin-deficient glia in both the PNS and CNS [194, 310]. However the role that integrins play in earlier glial development or in non-myelinating glia is less clear. Moreover while components of the adhesion complex are expressed in glia [311], the role of talin in vertebrate glial development has yet to be determined. . The complexity and array of integrin dimers [153, 312] complicates the investigation of integrin function in vertebrates.  In contrast, Drosophila has a simple family of  integrin subunits consisting of five alpha and two beta subunits [278].  The Drosophila  adult compound eyes, which originate from larval eye imaginal discs, are an excellent model to study many aspects of glial development, as well as glia-neuron interactions [125]. In particular the eye system allows for the detailed analysis of both glia-axonal, glia-glial and glia-ECM analysis from the earliest stages of glia development. The integrin associated focal adhesion kinase (FAK) has been suggested to genetically interact with βPS integrin and control optic stalk (OS) morphogenesis [313]. However whether integrins and other adhesion complex components are necessary for glia migration and development in the eye has yet to be determined.  Here we show that  Drosophila integrin mediated glia-ECM and glia-glia interactions are necessary for retinal glia development.  Specifically the loss of βPS integrin or talin compromises glial  migration into the eye disc.  Unexpectedly we found that loss of the integrin complex 97  leads to changes in glial organization within the OS paired with axon stalling in the OS and a failure of axons to enter the brain lobe.  98  3.2 Methods and materials 3.2.1  Fly strains and genetics  The following fly strains were used in this study: repo-GAL4 [93]; spg-GAL4 [279]; UAS-mCD8::GFP [282]; UAS-mCD8::RFP (a gift from Dr. Elizabeth Gavis); UAS-Dicer2 [284]; mys1 [286]; FRT19A,tubP-Gal80,hsFLP,w* [282]; repo-FLP [109]; GMR-myr-mRFP [314], UAS-p35 [315], UAS-λ-htl [316], P{PZ}GlirL82 (Gli-lacZ) [107]. The following GFP protein-trap insertions were used: Neurexin IV::GFP (NrxIV::GFP); integrin-linked-kinase::GFP (ILK::GFP) [287, 288]; multiple edematous wing::YFP (αPS1::YFP) and inflated::YFP (αPS2::YFP) [317]; talin::GFP (a gift from Dr. Guy Tanentzapf).  The following UAS-RNAi strains were used in our experiments:  UAS-βPS-RNAi(GD)(GD15002),  UAS-talin-RNAi(GD)(GD12050),  UAS-αPS2-RNAi  (GD44885) and UAS-αPS3-RNAi (GD4891; KK100949) are from the Vienna Drosophila RNAi Center [284]; UAS-βPS-RNAi(R1) (1560R-1), UAS-αPS2-RNAi(R1) (9623R-1), UAS-αPS3-RNAi(R1) (8095R-1) and UAS-talin-RNAi(R1) (6831R-1) are from the National Institute of Genetics, Japan.  RNAi experiments were carried out at 25°C with  UAS-Dicer2 plus UAS-CD8::GFP in both control (crossed with w1118) and experimental crosses with the exception of UAS-βPS-RNAi(GD), which was used without co-expression of Dicer2. For example, the full genotype for repo-GAL4 or ‘control’ is UAS-Dicer2; repo-GAL4,UAS-CD8::GFP/+. 99  3.2.2  Immunohistochemistry and imaging analysis  The following primary antibodies were obtained from the Developmental Studies Hybridoma bank developed under the auspices of the NICHD and maintained by The University of Iowa, Department of Biology (Iowa City): mouse anti-βPS (CF.6G11)[198] at 1:10, mouse anti-αPS2 (CF.2C7)[198] at 1:5, mouse anti-Repo (8D12) [289] at 1:50, mouse anti-chaoptin (24B10) [318], mouse anti-β-galactosidase (40-1a) [319] at 1:50. Other primary antibodies were: rabbit anti-αPS3 antibody (a gift from Dr. Shigeo Hayashi, Riken Center for Developmental Biology, Japan) [290] at 1:200, rabbit anti-HRP (Jackson ImmunoResearch) at 1:500, rabbit anti-Laminin-γ (LanB2) (Abcam) at 1:100. All the following secondary antibodies were used at a 1:200 dilution: goat anti–mouse Alexa568 and Alexa647, goat anti–rabbit Alexa 488, Alexa568 and Alexa647 (Molecular Probes/Invitrogen);  peroxidase-conjugated  goat  anti-mouse  IgG  (Jackson  ImmunoResearch). Dissection and fixation for immunofluorescence was performed according to standard procedures [92].  For integrin staining, larvae were fixed in 4%  paraformaldehyde in phosphate-buffered saline for 10 minutes at room temperature.  For  immunohistochemistry, eye discs were incubated with DAB (SIGMA FAST™ DAB (3,3′-Diaminobenzidine tetrahydrochloride) with Metal Enhancer Tablets, Sigma-Aldrich) for 30-60 minutes. For actin staining, all body wall muscles were removed before incubation with phalloidin Alexa568 (Molecular Probe) at 1:50. 100  Unless specified fluorescent images were obtained with a DeltaVision Spectris restoration system (Applied Precision), using a 60X oil immersion lens (NA 1.4), with 0.2 micron Z section increments.  The image stacks were subsequently deconvolved with  SoftWorx (Applied Precision), based on a measured point spread function obtained using 0.2 micron fluorescent beads (Molecular Probes) mounted in Vectashield (Vector Laboratories). Images were then exported to ImageJ, Photoshop and Illustrator CS4 for compilation.  For lower magnification images, images were captured with the  DeltaVision system, using a 20X air lens (NA 0.4) or with an Axioskop 2 system (Zeiss), using a 40X air lens (NA 0.75).  For DIC (differential interference contrast) images, the  pictures were captured using a 5X (NA 0.13) or a 20X air lens (NA 0.5) with an Axioskop1 (Zeiss).  3.2.3  Statistics  For statistical analysis, each independent replicate ‘n’ value represented the glial number from one eye disc or OS or eye disc plus OS from one half of the brain.  The  average glia number in each eye disc and/or OS was expressed as mean±standard deviation (M±SD).  The mean, standard deviation and standard error were calculated  and all graphic charts were made with Prism 5 (Graphpad Software).  All comparisons  were conducted with Kruskal-Wallis test with Dunn’s Multiple Comparison post-test. Significant differences were assessed at a 95% confidence interval. 101  3.3 Results 3.3.1  Integrins are expressed in both populations of glia in the larval optic stalk  In the late larval stages photoreceptor axons extend from the eye disc through the optic stalk (OS), a tubular structure connecting the eye disc and brain, into the brain lobe [123]. In conjunction, glia migrate from their origin in the OS into the eye disc [124]. Two glial subtypes are initially found in the OS, two central carpet glia (CG) and many surrounding perineurial glia (PG) (Fig. 20A).  The two CG are specialized  subperineurial glia (SPG) and display a tube-like structure along the OS (Fig. 20A, B, F). At later stages in the 3rd instar larva, photoreceptor axons from the developing eye disc will migrate through this tube [126] (Fig. 20G). Surrounding the two CG is a single outer layer of PG, an actively dividing population of glia (Fig. 20A). In the middle 3rd instar and later stages, both CG and PG migrate into the eye disc along the basal level of the disc epithelia and photoreceptors [126]. The PG migrate into the eye disc between the CG and the basal ECM.  At the distal end of the CG border  the PG are able to detect neuron derived FGF signals that trigger differentiation of the PG into wrapping glia (WG) [130].  Differentiated WG switch their migration direction and  travel back through the OS as they contact and ensheath the photoreceptor axons in the eye disc and OS (Fig. 20G).  A population of PG remains as a single layer of cells  surrounding the OS (Fig. 20G). To answer whether integrins are expressed by the glia in the OS and eye disc, we 102  detected integrin protein expression using antibodies to the single essential β subunit in Drosophila, βPS, encoded by the myospheroid (mys) gene.  At the early 3rd instar stage  (eL3), βPS integrin was detected in glia in the OS (Fig. 20B-D).  In the glia, βPS  integrin was concentrated in puncta similar to the distribution observed in the peripheral nerve [320].  The βPS puncta were found between the PG and the extracellular matrix  (ECM), between the PG and CG, and between the two CG (Fig. 20C-D).  At the later  wandering 3rd instar stage (wL3), βPS integrin expression persisted in glia (Fig. 20G). For example, βPS integrin labeling was seen in both the peripheral PG and central WG regions in the OS (Fig. 20G). Drosophila melanogaster has five alpha integrin subunits of which only αPS1-3 are homologous to vertebrate alpha integrin subunits [200].  Using immunolabeling and  alpha subunits endogenously tagged with YFP, we observed αPS2 and αPS3, but not αPS1, in the OS and eye disc.  In the early L3 OS, αPS3 integrin was predominantly  found at the PG-CG interface (Fig. 20E) and αPS2 integrin was mostly between the PG and ECM (Fig. 20E). To determine if the distribution of alpha and beta integrins is indicative of focal adhesions, we assayed for the presence of talin and found βPS integrin also co-localized with talin endogenously tagged with GFP (Fig. 20F).  These results  suggest that βPS integrin with specific alpha subunits are recruited to focal adhesions to mediate glia-ECM, glia-glia, or glia-neuron interactions.  103  104  Figure 20: βPS integrin and Talin expression in glial cells in the eye disc and optic stalk A) Repo immunolabeling (red) shows all glial nuclei in an early 3rd instar (eL3) optic talk (OS). The two carpet glia (CG) were labeled by SPG>CD8::GFP (green) and their nuclei (asterisks) are bigger than perineurial glial (PG) nuclei. DAPI (blue) labeled all nuclei in the OS and eye disc (ED). B-D) βPS labeling in an eL3 OS.  SPG>CD8::GFP (green) labeled two CG  (asterisks) and DAPI (blue) labeled all nuclei. The region in the yellow box was digitally expanded and shown in C.  The green line in B shows the position of orthogonal  sections in D. βPS integrin forms puncta located at the PG-ECM (arrowheads) and PG-CG (arrows) interface. E) αPS2 (green) and αPS3 (red) integrin subunits in an eL3 OS. αPS2::YFP was predominantly expressed in the outer PG (arrow) while αPS3 immunolabeling was predominant in the central CG (arrowhead). F) Co-localization of Talin::GFP (green) and βPS immunolabeling (blue) in the eL3 eye disc and OS. The CG were labeled by SPG>CD8::RFP (red). G) βPS immunolabeling in the wandering 3rd instar (wL3) eye disc and OS. repo>CD8::GFP (green) labeled all glia and GMR-RFP (red) marked photoreceptor cells and axons.  βPS integrin (blue) immunolabeling was found in the PG membrane  (arrowheads) and in the center of the OS (arrows). All panels are single 0.2 μm Z-sections except the projection in A. E-G were digitally expanded and shown in the smaller  panels.  Yellow boxes in  Scale bars: 10 μm in A,  B and big panels in E and F; 5μm in C and D, small panels in E and F, and right three panels in G.  105  3.3.2  Loss of βPS integrin in glial cells causes optic stalk morphological changes  βPS integrin has been shown to genetically interact with Fak56D and regulate PG distribution along the OS [313].  However, there is no evidence to show whether  integrins are also involved in other important aspects of glial activities, such as glial migration and differentiation.  Loss of function mutants in the βPS gene (mys) are  embryonic null, therefore to investigate integrin function in OS and eye disc development in 3rd star larvae, we took advantage of the RNA interference (RNAi) technique. To analyze the roles of integrin at later larval stages, RNAi experiments were conducted using repo-GAL4 to drive independent RNAi lines previously shown to knock down PS and talin [226, 320].  For PS integrin βPS-RNAi(R1) with Dicer2 and  βPS-RNAi(GD) without Dicer2 were used.  βPS-RNAi (GD) expressed with Dicer2 had  retarded animal growth that stalled in the 3rd instar.  In these larvae, glial proliferation  was inhibited within the OS (data not shown) making further analysis of the later roles of integrin in eye development difficult.  In the RNAi treated OS, PS integrin  immunolabeling was absent from the glia, while the integrin complexes within the eye disc epithelia were not disrupted (Fig. 21). Two independent RNAi lines for talin, talin-RNAi(GD) and talin-RNAi(R1) (both with Dicer2), were also used and disrupted the βPS puncta within the OS glia (Fig. 21C). In the βPS-RNAi OS, we observed morphological changes such that the anterior end closest to the eye disc became thicker compared to controls (Fig. 22). 106  This phenotype  was observed in both early (Fig. 21B) and wandering 3rd instar (Fig. 22B-C) larvae and was consistently observed in both independent lines (Table 4). In addition we were able to rescue this phenotype by co-expressing a UAS-PS transgene along with the UAS-PS-RNAi line (data not shown).  The thicker OS phenotype was also observed  in repo>talin-RNAi (Fig. 21C, 3D) and at a frequency similar to the βPS-RNAi lines (Table 4).  This suggests that the OS phenotype is due to disruption of the integrin based  focal adhesion complex in glia.  107  Figure 21.  Knock down of integrin and OS morphological changes in early 3rd  instar. repo-GAL4 was used to express βPS-RNAi(R1) and talin-RNAi(GD) in all glia. UAS-CD8::GFP (green) labeled glial membranes and DAPI (blue) marked nuclei. Images were projections of 0.2 μm Z-stacks.  In a control OS (A), βPS integrin  immunolabeling was detected as puncta in the eye disc and OS.  These puncta were  absent in βPS-RNAi OS (B) and greatly decreased in talin-RNAi OS (C) (arrows), but still present in the eye disc epithelia (double arrows).  Note that PG were mostly  detected in the anterior half (close to ED) and barely found in the posterior half (arrowheads) of RNAi OS.  108  109  Figure 22. OS morphological changes at wandering 3rd instar. repo-GAL4 was used to express different RNAi lines in all glia with UAS-CD8::GFP (green) to label glial membranes.  Dicer2 was co-expressed in A, C and D.  Glial nuclei  and photoreceptor cells were labeled by Repo (red) and HRP (blue) immunolabeling respectively. A-D) When comparing with the control (A), repo>βPS-RNAi and repo>talin-RNAi had thicker OS (B-D, arrowheads) in which Repo-labeled glial nuclei often aggregated at the anterior half of OS and CD8::GFP labeled glial membrane extended to the fine posterior ends of the thicker OS (B-D, arrows). E) Expression of an activated FGFR, λ-htl resulted in extremely thick OS (double arrows), which contained many more glia than the control and other genotypes.  Glia  migration into the eye disc was reduced (arrows) but photoreceptor axons did not stall in the OS. F) When βPS-RNAi (GD) was co-expressed with λ-htl, glial over-proliferation in the OS was still observed (double arrows).  Note that glial migration into the eye disc was  completely blocked and photoreceptor axons stalled within the OS (arrowheads). All images were taken using a 40X objective with an Axioskop 2 system (Zeiss).  .  110  Optic stalk phenotype Genotypes repo-GAL4 repo>βPS-RNAi(R1) repo>talin-RNAi(GD) repo>talin-RNAi(R1) *repo-GAL4 *repo>βPS-RNAi(GD) *repo>λ-htl *repo>λ-htl> βPS-RNAi(GD) *repo>λ-htl> talin-RNAi(R1) repo>αPS2-RNAi(GD) >αPS3-RNAi(R1) repo>αPS2-RNAi(R1) >αPS3-RNAi(GD)  Axonal phenotype classes  Normal  Thicker  Thinner  Total  Normal  Mis-targeting  Stalling in OS  Stalling in ED  Total  29 0 0 0 22 3 0  0 56 36 45 0 41 25#  0 3 1 3 0 0 0  29 59 37 48 22 44 25  40 18 13 25 32 14 13  0 6 21 15 0 13 21  0 16 18 7 0 5 0  0 5 9 3 0 10 0  40 45 61 50 32 42 34  0  40#  0  40  0  7  50  5  62  0  28#  0  28  3  13  37  1  54  -  -  -  -  32  4  1  2  39  -  -  -  -  5  2  5  8  20  Table 4. Summary of glial and neuronal phenotypes. All the genotypes that were examined in the text for the optic stalk morphologic changes and photoreceptor axon outgrowth and targeting phenotypes were listed. specified.  All RNAi and control crosses were done with co-expression of UAS-CD8-GFP and at 25 °C unless  UAS-Dicer2 was co-expressed in the crosses that were not marked with (*).  were much larger than the optic stalks in other genotypes.  Note that the optic stalks marked with (#)  Abbreviations: ED: eye disc; OS: optic stalk. 111  3.3.3  βPS integrin is required for glial organization and migration in the optic  stalk Two glial aberrations were observed that corresponded to the morphological changes in the OS observed in both early and wandering 3rd instar larvae (L3).  First, PG nuclei  tended to aggregate in the anterior half of the thicker OS, unlike the roughly equal distribution in control OSs (Fig. 21A, 3A, 4C).  This left the posterior end (close to the  brain) with a fine and skinny appearance (Fig. 21B-C, 3B-D, 4D). The posterior end contained the CD8::GFP labeled glial membrane that marked the projection of the CG along the full extent of the OS.  Second, in the radial direction the PG became  multi-layered around the thicker OS instead of creating the single cell layer seen in controls (Fig. 23C-D). We observed that more glia were located in RNAi expressing OSs than in controls. For comparison, glial numbers were counted in the wandering L3 stage with 11-15 rows of ommatidia in the eye disc (Fig. 23A).  On average RNAi treated OSs contained  36%-56% more glia than controls (Table 5). The presence of extra glia within the OS could be due to a reduction in migration into the eye disc and we observed that RNAi expressing eye discs had 33%-54% less glia than controls (Table 5).  To better show the  relative distribution of glia in the eye disc versus the OS, the ratio of glia in the OS was compared the entire population in both the OS plus eye disc (Fig. 23B). This ratio significantly increased from 0.22 in controls to 0.41- 0.50 in RNAi expressing larvae 112  (Table 5).  Thus, our data suggested that βPS integrin and talin may play a role in glial  migration from the OS into the eye disc and the changes to glial number and their distribution led to a thickening of the OS.  113  114  Figure 23. Knockdown of integrin and talin compromised glial migration and organization. A) Quantification of glial number separately in the ED and OS in repo>Dicer2 (Control),  repo>βPS-RNAi(GD)  (*βPS-RNAi(GD)),  repo>Dicer2>βPS-RNAi(R1)  (βPS-RNAi(R1)) and repo>Dicer2>talin-RNAi (talin-RNAi(GD) and (R1)). Glial number was scored by Repo immunolabeling at wandering L3 with 11-15 rows of ommatidia. Bars represent the Standard Deviation. B) The ratio of glia in the OS (GliaOS) compared to glia in the OS plus eye disc (GliaOS+ED). Bigger ratios in RNAi larvae suggested that relatively more glia were in the OS.  Lines indicate the mean ratio and standard error of the mean. All comparisons had  P<0.05 (one star), P<0.01 (two stars) or P<0.001 (three stars) as indicated. C-D) High resolution images of control and βPS-RNAi OS. Glial membrane and nuclei were labeled by CD8::GFP (green) and Repo (red) immunolabeling respectively. Photoreceptor cells and axons were immunolabeled using anti-HRP. Lower magnification images, taken using a 40X objective, show global views of the eye disc and OS (left). The regions in the red boxes are shown at high magnification (60X objective) with single 0.2 μm sections (middle) and projections of the entire Z-stack (right). The arrowheads in C and D point to single PG layer or multiple PG layers. The arrow in D indicates the fine posterior end of the thicker OS.  Dashed lines show the middle point of each OS.  Asterisks highlight the two CG nuclei.  Scale bars are 5 μm.  115  repo-GAL4  repo>βPS-RNAi(GD)  repo>βPS-RNAi(R1)  repo>talin-RNAi(GD)  repo>talin-RNAi(R1)  192  128  106  89  93  21  48  28  30  32  6  14  6  6  7  59  86  80  87  92  12  27  27  28  24  3  8  6  6  5  0.223  0.414  0.424  0.496  0.505  0.044  0.111  0.109  0.115  0.109  0.012  0.032  0.024  0.024  0.023  13  12  21  23  22  Genotype Glia in ED  Glia in OS  OS  Glia /Glia  OS+ED  N  Mean SD SE Mean SD SE Mean SD SE  Table 5. Quantification of glia distribution in the eye disc and optic stalk. Glia were counted using Repo immunolabeling in the wL3 eye disc and optic stalk with 11-15 rows of ommatidia. UAS-Dicer2 was co-expressed in all genotypes except repo>βPS-RNAi(GD). mean .  116  A  SD: standard deviation; SE: standard error of the  3.3.4  Loss of βPS integrin in glial cells causes axonal migration defects  To address how neurons responded to the glial changes in loss of function of βPS integrin and talin, an antibody to Chaoptin (mAb 24B10) was used to label photoreceptor cells and their axons.  In control larvae, photoreceptor axons pass through OS and target  to specific layers in the optic lobe (Fig. 24A-B).  Knock down of βPS integrin in the glia  resulted in a range of photoreceptor axon phenotypes, which were grouped into three categories: i) mis-targeting in the brain, ii) stalling in the OS, iii) stalling in the eye disc (Fig. 24; Fig. 25) (Table 4).  The axonal phenotypes were also observed when  repo-GAL4 was used to knock down both PS2 and PS3, or talin in glia (Fig. 25) (Table 4), but not observed when GMR-GAL4 was used to knockdown integrin specifically in the photoreceptor cells (data not shown). In the mis-targeting phenotype, all or most of photoreceptor axons successfully passed through the OS but failed to terminate at correct regions in the brain and formed highly disorganized patterns (Fig. 24C,C’).  These data suggested that the reduction in  βPS and talin in the glia strongly affected photoreceptor axon guidance in the brain. In the two stalling phenotypes, photoreceptor axons failed to exit the eye disc (Fig. 24D,D’; Fig. 25B, B') or the OS (Fig. 24E, E’; Fig. 25C, C').  Defects in axonal  mis-targeting and lacking of photoreceptor innervation were also observed in the adult RNAi optic lobe (Fig. 26), suggesting that these neuronal defects were not recovered during later developmental stages. 117  Given the distribution of PS in different glial layers of the OS, we tested the contribution of each layer in mediating the glia and neuronal phenotypes.  However,  when drivers specific for each glial layer (C527-GAL4 and 46F-GAL4 for PG, SPG-GAL4 for CG, and MZ97-GAL4 and Gli-GAL4 for WG) were used to express either PS or talin RNAi, none generated the glial or neuronal phenotypes observed with repo-GAL4 (data not shown).  These results suggested it was the disruption of the  integrin complexes in a combination of different glial layers that contributed to the observed phenotypes.  These observations were confirmed when we analyzed somatic  MARCM clones of a null mutation in the PS gene, myospheroid (mys1).  We had  successfully used the MARCM system to obtain somatic clones of mys1 in the PG of the peripheral nerve previously (Xie and Auld, 2011).  Using this approach in the optic  nerve, we obtained multiple mys1 clones in the PG and the rare clone for one of the two CG (never both at once).  The mys1 clones observed in the glia of the OS and eye were  generally limited in size, and we failed to observe either the glia or neuronal phenotypes seen with global downregulation of PS in all glia using RNAi (data not shown). However it was impossible to obtain large clones that simultaneously involved both the CG and the PG cell types, further suggesting that the loss of PS or Talin in both glial layers is necessary for the observed phenotypes.  In support of this, when we knocked  down PS2 or PS3 individually in all glia using repo-GAL4 driven RNAi, we observed only rare axonal phenotypes (data not shown). 118  However when we knocked down both  PS2 and PS3 in combination using repo-GAL4, we observed similar axon stalling phenotypes as with PS and talin RNAi (Fig. 25A, A') (Table 4).  These results suggest  that the axonal phenotypes we observed were due to the combined loss of integrin complexes from multiple glial layers most likely the PG and CG.  119  120  Figure 24: Photoreceptor axon migration defects observed with integrin loss in glia. A monoclonal 24B10 antibody was used to label photoreceptor cells and their axons. Higher resolution images from those areas highlighted in the red boxes are also shown. A-B) In control larvae, photoreceptor axons exited the eye disc (ED), passed through the OS and terminated in the lamina (la) and medulla (me) layers (A’, arrowheads) in the brain lobe (Br). C-E’) In repo>βPS-RNAi larvae, neuronal defects were observed where the photoreceptor axons formed a disorganized pattern in the brain lobe (C and C’, arrow), failed to exit the eye disc (D and D’, arrowheads) or stalled in the OS (E and E’, arrowheads). F&F’) In repo>λ-htl, photoreceptor axons managed to penetrate the OS but displayed hyperfasciculation defects in the brain (arrows). Note that the OS was much larger and axons more separated than normal (arrowhead). (G-H’) When βPS-RNAi(GD) was co-expressed with λ-htl in glia, the OS remained swollen and photoreceptor axons were stalled in the eye disc and OS (arrowheads). Note that Bolwig’s nerve (BN) was also present in the OS in all stalling phenotypes and remained unaffected (D’, E’, G’, H’). Panel B was captured using a 5X objective with an Axioskop 1 (Zeiss). were captured using a 20X objective.  121  Others  Figure 25. Photoreceptor axon migration defects in alpha integrin and talin RNAi. A monoclonal 24B10 antibody was used to label photoreceptor cells and their axons. Axonal stalling phenotypes were observed in wandering 3 rd instar larvae of repo>αPS2-RNAi>αPS3-RNAi and repo>talin-RNAi.  Panel A, B and C were captured  using a 20X air lens and digitally condensed to 50% to show the complete structure. Regions in red boxes are shown in their original size in respective (’) panels.  122  Figure 26. Photoreceptor axon targeting defects in RNAi adults. A monoclonal 24B10 antibody was used to label photoreceptor axon in adult fly eye-brain complexes.  In controls (A), 24B10 labeled photoreceptor axons terminated  properly in the optic lobe (OL).  Note that left panel shows axonal termination in both  lamina (La) and medulla (Me) layers and only medulla was preserved in the right panel. In RNAi flies (B&C), different neuronal defects were observed, which included disorganized axons termination (arrowheads in B&C) and lack of photoreceptor innervation (arrows in B&C) in the optic lobe. with the abbreviated axonal termination.  Smaller optic lobes were seen associated  Dashed lines show the boundary between the  optic lobe and midbrain (MB).  123  3.3.5  Extra glia within the OS does not block axon outgrowth  To further characterize the axon stalling phenotype, fluorophore conjugated phalloidin was used to label F-actin in the axons stalled in the OS (Fig. 27). In contrast to the control (Fig. 27A), strong phalloidin labeling was detected in the stalling region (Fig. 27B), suggesting the region may include actin-rich growth cones. Since the axonal defects were caused by loss of glial integrins and correlated with the thick OS phenotype (Table 4), we then determined whether extra glia in the OS was sufficient to block axonal penetration.  It has been shown that heartless (htl), one of the  two Drosophila fibroblast growth factor receptors (FGFRs), controls the number of glia in the OS [130].  A constitutively active form of heartless, λ-htl was expressed in the glia  and as reported previously (Franzdottir et al., 2009), all OS (n=25) became extremely large and contained hundreds more glia than normal (Fig. 22E).  However,  photoreceptor axons successfully penetrated most of these extremely thick OS (Table 4) (Fig. 22E; Fig. 24F,F’), suggesting that extra glia do not necessarily prohibit axonal outgrowth through the OS. We then used repo-GAL4 to co-express βPS-RNAi(GD) or talin-RNAi(R1) with λ-htl.  The co-expression of βPS-RNAi(GD) and λ-htl resulted in a giant OS with many  extra glia (Fig. 22F).  In this case axonal outgrowth from the eye disc or into the OS was  inhibited in the majority of OSs (Table 4) (Fig. 22F; Fig. 24G-H’).  These results  confirm that it is the loss of βPS that results in a block in axonal penetration of the OS. 124  Moreover, we observed that when βPS-RNAi(GD) was co-expressed with λ-htl, glia failed to migrate into >50% of the eye disc observed (n=25/40) (Fig. 22F),  The  complete failure of glial migration into the eye disc was never observed when λ-htl was expressed alone (n=0/25) and with a low penetrance when βPS-RNAi(GD) was expressed alone (without Dicer2) (n=1/26).  This synergistic interaction suggested that βPS  integrin and a proper level of FGFR activity are required for normal glial migration in the eye disc and OS.  125  126  Figure 27. Axons stall near PG displaced into the center of the optic stalk. A-B) Phalloidin Alexa568 (red) was used to label F-actin in control (A) and repo>βPS-RNAi (B) eye disc and OS.  Glial membranes were labeled with  repo>CD8::GFP (green) and all cell nuclei were marked with DAPI (blue).  Note that  the axonal stalling region (arrowheads) was highlighted with concentrated phalloidin labeling and a posterior glial cap (marked with the double ended arrow) in the RNAi OS. C-E) γ-Laminin (Lam-γ, green) and Gli-lacZ (β-gal, blue) are markers for PG and differentiated WG in control (C) and repo>βPS-RNAi (D-F) wandering L3 OS. glial membranes were labeled using repo>CD8::RFP (red).  All  Regions in white boxes  were digitally expanded and shown in the small panels to the right. C) In the wild type eye disc and OS, Lam-γ labeling was predominately found in the outer ECM and PG (double arrows), but not in the inner WG which were distinct for their long membrane processes and expressing Gli-lacZ (arrowheads). D-F) In the repo>βPS-RNAi OS, Lam-γ was still detected in the ECM (double arrows) and surface PG.  Moreover Lam-γ positive glia (D-E, arrows) were observed in  the central region of the OS along with the WG (arrowheads) in OS that had penetrated axons (D) or stalled axons (E). In some stalling OS, Gli-lacZ labeling was found both in WG in their normal position (F, arrowheads) and in glia located in the glial cap posterior to the axonal stalling regions (F, arrows). All panels are single 0.2 μm Z-sections. panels and 5 μm in the smaller right panels.  127  Scale bars are 10 μm in the larger left  3.3.6  Aberrant glia deposition is observed in the thicker optic stalk  Overall distribution of glia was significantly changed in the OS with reduced βPS. As shown in Fig. 27B, when photoreceptor axons were stalled in the OS, a group of glia aggregated and appeared to form a cap-like structure posterior to the stalling region (further away from the eye disc towards the brain lobe). To answer how these glial changes resulted in axonal migration defects, we examined molecular markers that labeled specific OS glial subtypes.  Because of the lack of a good PG marker, we  examined the distribution of Laminin previously shown to be produced by the PG and deposited into the ECM in the peripheral nerve [320].  In control OSs (n=10), γ-Laminin  (LanB2) immunolabeling was predominantly observed in the ECM that surrounded the OS, as well in the cytoplasm of the PG (Fig. 27C).  This suggested that the OS PG like  the peripheral nerve PG are responsible for laminin production. In the repo>βPS-RNAi and repo>talin-RNAi OS, we observed laminin positive glia in the central regions of the OS in close proximity to photoreceptor axons (Fig. 27D-E). These laminin positive glia were different from normal WG in two ways. not have long bipolar membrane protrusions, a characteristic of WG. not express Gli-lacZ, a WG marker in the OS and eye disc.  First, they did  Second, they did  The Gli-lacZ reflects the  expression of the Gliotactin gene [107], which is expressed by the CG in the OS at stages prior to photoreceptor formation (data not shown).  At later stages beta-galactosidase  (Gli-lacZ) labeling was detected in both the CG (Fig. 27C, F) and the WG (Fig. 27C-F). 128  For these two reasons, the laminin positive glia were likely PG that had invaded into the central region of the RNAi OS.  We also observed a mislocalization of WG in the RNAi  OS and observed some glia in the glial cap that expressed the Gli-lacZ (Fig. 27F). Gli-lacZ labeling was limited to the regions immediately posterior to the stalled axons, suggesting that the Gli-lacZ positive glia were most likely to be WG that migrated beyond stalled axons and lost their association with the photoreceptor axons.  These  results lead to the question of how PG invasion happens in βPS and talin RNAi larvae.  3.3.7  Loss of integrins compromises carpet glial barrier in the optic stalk  In the normal OS, the PG are physically separated from the internal photoreceptor axons and the WG by the glial sheath formed by the two intervening CG [130].  Our  results suggest that PG invasion into the center of the OS occurs due to a disruption of the cellular barrier created by the CG.  In the absence of a global membrane marker for the  CG, we assayed the integrity of the CG sheath and changes to CG morphology using the transmembrane protein NeurexinIV endogenously tagged with GFP (NrxIV::GFP). NeurexinIV is one of the core structural proteins of septate junctions (SJs) and is specifically expressed in the subperineurial glia (SPG) to generate the blood-brain barrier [108, 109, 126].  In the control early L3 OS (n=6/7), NrxIV::GFP localized to two  condensed lines along the length of the OS (Fig. 28A-B) found between the two CG and extended to the eye-disc end of the OS where a NrxIV::GFP positive end-foot was seen 129  (Fig. 28B), suggesting that the septate junctions extend the entire length of the OS with the CG sheath. The organized NrxIV::GFP pattern was disrupted in the early L3 repo>βPS-RNAi OS. The majority of repo>βPS-RNAi(R1) OS (n=6/8) displayed a disorganized NrxIV::GFP localization in which NrxIV::GFP failed to form condensed lines in the eye-disc end of the OS and the NrxIV::GFP terminated in the middle of the OS (Fig. 28D).  Interestingly,  the NrxIV::GFP distribution seemed to be normal in the posterior half of the OS, where it was condensed and formed two lines associated with the tube created by the CG (Fig. 28D). The changes to NrxIV::GFP after knockdown of βPS integrin was not only restricted to the OS.  Similar NrxIV::GFP disorganization was also seen in the peripheral nerves of  repo>βPS-RNAi(GD) (Fig. 28F), suggesting that βPS integrin is required globally by Drosophila glia to trigger sheath and SJ formation. The OS phenotypes were observed in early L3 stages prior to photoreceptor axon outgrowth suggesting that the CG failed to wrap the anterior end of the OS allowing the PG to enter into the center of the OS and contact Bolwig's nerve (Fig. 29A).  By later  3rd instar stages, this suggests that the presence of ectopic PG perhaps in combination with the WG generates a glial cap that blocks photoreceptor axon outgrowth (Fig. 29C).  130  131  Figure 28. Carpet glia defects observed with integrin knockdown. A NrxIV::GFP was used to label the septate junctions (SJs) in early L3 OS (A-D) and wandering L3 peripheral nerves (E-F).  Most panels are projections of 0.2 μm Z-stacks  except that the right panels in A-D are orthogonal sections of the Z-stacks and their positions are marked by green lines. A-D) Glial membrane was labeled by repo>CD8::RFP (red) and Bolwig’s nerve by 24B10 immunolabeling (blue) in the OS.  In a control OS (A-B), NrxIV::GFP located to  two lines of SJs (arrows) along the OS and formed an endfoot at the anterior end of the OS (arrowhead).  The SJs were formed between two CG, which closely associate with  Bolwig’s nerve at this stage.  Note that NrxIV::GFP labeled SJs were also found  between epithelial cells in the ED. C-D) In a repo>βPS-RNAi(R1) OS, NrxIV::GFP concentrated lines stopped in the middle of the OS (arrowhead) and became diffuse in the anterior glial membrane (double arrows). The CG still formed a tube-like structure in the posterior half of the OS in which the NrxIV::GFP condensed into two lines as normal (arrow in D).  Note that PG tended  to aggregate at the anterior half of the RNAi OS. E-F) Repo and HRP immunolabeling were used to label glial nuclei and axons in the peripheral nerve, respectively.  In the control nerve (G), the SJs as shown by a single  line of NrxIV::GFP are formed between the SPG membranes.  Similar NrxIV::GFP  disruption was observed in the repo>βPS-RNAi(GD) peripheral nerve (H) as in the repo>βPS-RNAi(GD) OS . Scale bars are 10 μm in left panels in A-D and all panels in E&F, and 5 μm in right panels in A-D.  132  133  Figure 29. Model of integrin disruption of glial organization in the OS. A model of integrin complex disruption of the glia used to show a wild type (WT) OS (A&B) and a repo>βPS-RNAi thicker OS with axonal stalling (C-D) at early and wandering 3rd instar stages.  For a detailed explanation of the diagrams refer to the  Discussion section.  134  3.4 Discussion 3.4.1  Integrins play multiple roles in the glia in the optic stalk and eye disc  We showed that glia in the OS express integrin complexes that play a role in the development of both the glia and axons of the eye disc and optic stalk.  First, βPS  integrins are located to the glial membrane and concentrated in puncta that are associated with Talin.  Second, integrins are found between the PG and ECM, plus the different  glial layers PG-CG, CG-CG and possibly WG-axon interfaces.  Third, αPS2 and αPS3  integrins are found differentially concentrated in the peripheral and central regions in the OS.  Using RNAi mediated knockdown of the integrin complex, we determine these  integrin complexes play important roles in OS glial development.  Specifically we  observed changes to both the PG and CG in early 3rd instar larvae with associated axonal phenotypes by the later wandering 3rd instar. However both the glia and axonal phenotypes observed were dependent on simultaneous depletion of the integrin complex from both the PG and CG.  PG and CG  morphology appeared normal in the βPS null MARCM clones we observed in these glia subtypes or when the PG or CG specific drivers were used to down-regulate βPS integrin or talin (data not shown). This suggests that the glial and axon phenotypes we observed require the presence and function of integrin complexes in both the PG and CG.  This  possibility is supported by our observations that only simultaneous knockdown of both 135  PS2 and PS3 in all glial layers triggered the same phenotypes. In this regard the loss of PS integrin or talin caused the PG to aggregate at the anterior half prior to photoreceptor axon outgrowth.  The PG became multiple-layered  instead of single-layer uniformly distributed along the OS.  These results suggest that  PG make integrin mediated associations with both the overlying ECM and the underlying CG.  The αPS2/βPS heterodimer binds ligands containing the tripeptide RGD sequence  [201, 234] and αPS3/βPS can bind laminins [203, 204] so both could mediate adhesion of the PG to the overlying ECM.  Regardless the disruption of the integrin complex in  both layers was necessary to reduce glial migration into the OS.  As the PG migrate into  the eye disc between the CG and the basal ECM [126], this suggests a role for the integrin complex in both layers to mediate these processes. PG migration was an accumulation of glia in the OS.  The end result of reducing  When a constitutively active form  of the FGF receptor heartless was co-expressed this migration phenotype was enhanced and frequently blocked the migration of all PG, supporting the idea that integrins play a role in directing glial cell migration in the OS.  In vertebrates, prior in vitro studies have  shown that integrins are involved in astrocyte migration [321], oligodendrocyte precursor migration [162] and Schwann cell migration on various extracellular matrix molecules [196]. Though in the optic, integrin mediated migration appears to involve both the surrounding ECM and the underlying CG. The loss of integrins from both the PG and CG was necesary to disrupt CG 136  morphology and the NrxIV::GFP pattern.  It is possible the CG interact directly with the  PG or through an intermediate ECM as the PG do express and secrete laminins. Knock down of the integrin complex also disrupted the NrxIV pattern in the SPG of the peripheral nerve suggesting that maintenance of the SJ domain or the SPG sheath requires integrin-mediated adhesion in all SPG glia. Surprisingly integrins seem to play a limited role in the late onset wrapping glia (WG) in the eye disc and OS.  Using the pan glial repo-GAL4 diver or WG specific drivers  (MZ97-GAL4 or Gli-GAL4) to knock down βPS, normal WG with long bipolar membrane processes were still found in the eye disc and OS following axons (Fig. 27 and data not shown).  We also observed that WG clones of mys1 (βPS) MARCM had similar  morphologies as control clones (data not shown).  This suggests that once the WG have  differentiated and begun their developmental program, integrin signaling is not required at these stages. However integrins could be required in pupal or adult stages to stabilize the glial wrap.  3.4.2  Glial organization is important for axonal outgrowth in the optic stalk  Loss of the integrin complex in both PG and CG resulted in a failure of photoreceptor axons to exit the eye disc, to penetrate the OS or to targeting the correct region within the CNS.  Dominant negative Ras1 blocks glial cell migration when  expressed in optic stalk glia with the result that photoreceptors axons stall in the eye disc 137  [124]. The phenotype was interpreted as evidence that photoreceptor axons required physical contact with the glia to exit the eye disc.  However, the mechanism might not  be true for the stalling phenotype observed with knockdown of the integrin complex.  In  these tissues, glia were still present in the eye disc and around the axonal stalling region. Our axonal stalling phenotypes might be due to a different mechanism that involves disruption of the integrin complexes in both the PG and CG layers.  In our model  (Figure 8), it is the simultaneous aggregation of the PG due to reduced migration and the disruption of the CG sheath that normally creates a tube separating the PG from the axons in the OS [126, 130]. In the CG, the SJ marker NrxIV::GFP is disrupted suggesting a failure of the SJ domain or a failure of CG to establish and maintain ensheathment.  We  favour the later given the diffuse NrxIV pattern throughout the CG processes when the integrin complex is knocked down.  The combination of PG accumulation and the  disrupted CG sheath results in some PG moving in the OS center as suggested by the presence of internal Laminin expressing glia. Those Gli-lacZ expressing wrapping glia observed beyond the axonal stalling regions may represent WG that have migrated beyond the axons.  It is unclear whether the invading PG are still competent to respond  to the axonal signals that normally trigger PG to WG differentiation.  Thus an  alternative explanation for those posteriorly localized Gli-lacZ positive glia is that they might be differentiated from the laterally migrating PG. It is unclear how the PG invasion is related to the axon stalling. 138  One possibility is  that they might form a ‘plug’ that either physically and/or chemically blocks the progress of the photoreceptor axons through the CG tunnel.  Although the internalized PG were  observed in OS with and without axon stalling, the timing of invasion would be critical. Earlier invasion prior to axonal outgrowth may provide a greater opportunity to form an effective plug.  Invasion at later stages after the pioneer photoreceptors have  successfully navigated through the OS may not be sufficient to block further axon migration. In summary we have shown that glia in the OS and eye disc recruit integrins and talin to receive external signals that are important for PG migration and organization, and CG septate junction formation.  The combined impact of integrin complex on glial  development is critical for proper axonal outgrowth through the OS and termination in the brain.  139  Chapter 4  General conclusion  4.1 Summary of findings Glial development is controlled and modulated by many extrinsic factors including integrin mediated glia-ECM interactions.  However integrins’ roles in the glia have not  been completely revealed because of the existence of a large number of integrin subunits and difficulties of conducting in vivo experiments in vertebrates.  Drosophila  melanogaster has comparable glial types to vertebrate glia and a much smaller integrin family.  The primary aims of my PhD work were to determine the in vivo functions of  integrins in the Drosophila glia and to learn how the development of a specific type of glia is coordinated with the development of other glial cells and neurons. In my PhD dissertation, I have explored the distribution and examined the requirement of integrins in two groups of nerve associated glia, in the larval peripheral nerve and the optic stalk.  Both groups of glia contain three morphological and  molecular distinct glial subtypes. expressed by these glia.  Two Drosophila integrins, αPS2βPS and αPS3βPS are  Both integrins share a similar spatial distribution in the  peripheral nerve and optic stalk, with αPS2βPS integrin mostly found in the outermost perineurial  glia  (PG)  and  αPS3βPS  predominantly  subperineurial/carpet glia (SPG/CG) and wrapping glia (WG). 140  detected  in  the  inner  Loss of function experiments show that Drosophila integrins are necessary for normal glial development. nerve and optic stalk.  Integrins are required differently, however, in the peripheral  In the peripheral nerve, integrins seem dispensable for early  embryonic peripheral glia migration and their role in later glial proliferation and differentiation is unclear.  However, they are important for PG and WG sheath formation  and septate junction organization in the nerve.  The integrin-mediated PG wrapping  probably depends on the NL because removal of the ECM from the 3rd instar larval nerve generates a PG phenotype similar to the loss of integrins in the PG (Chapter 2). In the optic stalk, integrins are critical for PG proliferation, migration and organization in the optic stalk, as well as the formation of the septate junctions in the CG. Removal of integrins from glia also resulted in abnormal PG and WG depositions, and photoreceptor axonal outgrowth stalling in some eye discs and optic stalks (Chapter 3).  4.2 Significance 4.2.1  Insights on using Drosophila to study integrin functions in the glial  development My studies show that Drosophila integrins, particular αPS2βPS and αPS3βPS share similar properties as their vertebrate homologs in the glia.  First, Drosophila integrins  form adhesion complexes with Talin and ILK, which are broadly distributed in different 141  glial subtypes.  Some of the integrin complexes are located between the glia and ECM,  suggesting that Drosophila integrins, similar to vertebrate integrins, are used by glia as ECM receptors. development.  Second, Drosophila integrins are involved in multiple aspects of glial Some of them, such as glial proliferation, migration and axonal  ensheathment have been shown to be integrin dependent and be regulated by different integrin down-stream signalling pathways in the vertebrate nervous system [35, 162, 170, 171, 193, 321-325]. Thus basic integrin functions in glial development appear to be conserved in Drosophila and vertebrates, laying the foundation that Drosophila can be used as a model organism to answer how glial development is regulated by integrin mediated ECM interactions. My studies also implicate integrin functions that have not previously been shown in the vertebrate glial development.  For example, Drosophila integrins are clearly located  between different glial layers where no ECM has been detected, suggesting that Drosophila integrins may recognize cell surface components other than ECM ligands. In the Drosophila peripheral nerve, integrins are required for the WG to initiate and maintain proper membrane protrusions between axons.  Non-myelinating SC in the  vertebrate PNS ensheath axons in a similar matter, in which process laminins seem to be important [197].  Although it is not confirmed, my findings in Drosophila integrins and  other studies on laminins provide clues that integrins might mediate non-myelinating SC wrapping in the vertebrates.  Drosophila integrins might participate in developmental 142  processes that are unique to the fly glia.  For example, integrin absence in the glia  causes disruption of the septate junctions, which are mostly found between the invertebrate cells but are functionally replaced by the tight junctions between vertebrate cells.  4.2.2  Implications of integrin mediated glia-glia interactions  Developing glia receive regulatory signals from extrinsic resources, such as the ECM, and neurons.  However, it remains largely unknown how glial development is modulated  by other glial cells.  By studying integrin functions in the Drosophila glia, I found that  integrin mediated glia-glia cross-talk probably has an important role in glial development. In the peripheral nerve, integrins in the internal SPG and WG recognize ligands that are present between glial layers or on the surface of opposing glia membrane. Transmission electron micrograph (TEM) of the nerve shows that WG membrane can form adhesive structures at certain points with opposite membrane from either the same WG or from the SPG (unpublished data).  It is unclear whether these glia-glia adhesions  are mediated by integrins or other adhesion molecules.  However integrin mediated  interactions between same or different classes of glia seem critical for glia to form proper sheaths in the peripheral nerve. A similar integrin mediated glia-glia interaction is seen in the optic stalk, where integrin complexes lie between the PG and CG. The removing of integrins from all glia 143  results in the disruption of the septate junctions in the internal CG, suggesting that integrin mediated PG-CG interactions are important for the barrier formation in the optic stalk.  Integrin is also found to promote glial proliferation in the optic stalk and this  function seems to be independent to those mostly common integrin downstream partners including Talin, ILK and FAK. receptors.  A question arises whether integrins signal through other  Glial proliferation has been shown to be regulated by the Drosophila FGFR  Heartless (Htl) and its ligand Pyramus, which is produced by carpet glia.  It has long  been known that vertebrate integrins can cross talk with other growth factor receptors [326].  For example, ανβ3 integrin is shown to bind to FGF1 and FGF2 in endothelial  cells [327, 328] and the presence of FGFRs on cell surface is dependent on activation of certain integrins in the same cell [329].  My work in section 3.3.5 implicated a potential  genetic interaction between integrin and htl in glial migration.  It is still unclear whether  and how cross-talking between integrins and FGFRs occurs in the glia and regulates glial proliferation and differentiation.  4.2.3  Contribution to the studies of coordination between glial and neuronal  development Normal nervous system development requires both glia and neurons.  In the  Drosophila developing visual system, photoreceptor axons contact different types of glia prior to reaching their destinations in the optic lobe. 144  Some of these glia have been  shown to be required for proper axon outgrowth and termination.  For example, physical  contact with glia in the optic stalk is required for photoreceptor axons to exit the eye disc [124].  Glia in the optic lobe are necessary for photoreceptor axon target selections [94].  My studies show that integrin deficient glial cells can block and mislead photoreceptor axon outgrowth at various points on the way to their final targets. Inside the eye disc and optic stalk, integrin deficient glia fail to migrate and organize normally and they probably form a physical or chemical barrier to inhibit axonal extension.  In the brain,  photoreceptor axons terminate to wrong layers or completely different regions when integrins are absent in the glia, suggesting a guiding role of glia for the axons.  In summary, my studies confirmed integrins’ important roles in Drosophila glial development.  Integrins are not required equally in different glial subtypes.  They may  mediate glia-ECM, glia-glia and glia-neuron interaction through unidentified ECM or non-ECM ligands.  By removing integrins from glia, I also showed that glia were  critical for photoreceptor axons to pass through the optic stalk and terminate in the brain.  4.3 Future directions 4.3.1  Identification of glial integrin ligands  I have shown that integrins are consistently detected in both marginal (mostly 145  αPS2βPS integrin) and central (predominantly αPS3βPS integrin) regions of the peripheral nerve and optic stalk.  However their extracellular ligand identities have not  been completely determined. Vertebrate integrins bind both ECM and non-ECM ligands.  But in Drosophila, only  a small number of ECM proteins have been identified as integrin ligands, which includes two laminins [240, 245], tiggrin [234], thrombospondin [330] and tenectin [331].  Other  ECM proteins, such as Collagen IV [332] and tenasin [243] are present in the fly but not confirmed to be integrin ligands in vivo.  Some of ECM components (laminins, Collagen  IV and Perlecan) have been found in the neural lamella (NL) on the surface of the peripheral nerve and optic stalk.  Integrins, especially αPS2βPS integrin lying between  the PG and NL likely bind to some of these ECM components, such as Laminin.  This  idea is supported by the fact that degradation of the ECM and down-regulation of integrins in the PG share a similar glial wrapping phenotype in the peripheral nerve. However it is also important to note removal of the ECM gave phenotypes, such as glial death in the peripheral nerve (data not shown), that were not seen in integrin RNAi.  It  suggests that ECM components can also signal through receptors other than integrins, such as Perlecan and its receptor Dystroglycan [333, 334] and regulate alternative aspects of glial development. Another unsolved question is what the predominant internal αPS3βPS integrin recognizes.  These are two possibilities regarding the ligand identities of the internal 146  integrins.  First, although the ECM is predominantly observed in the NL in the electron  micrograph and Collagen IV-GFP and Perlecan-GFP are only limited to the neural lamella, it is still possible that low levels of ECM proteins, such as laminins might be deposited to specific domains between glial cells and neurons.  This idea is supported by  the fact that axons provide laminin-2 to regulate oligodendrocyte development in the vertebrate CNS [163, 172-174].  Unfortunately the laminin staining in the peripheral  nerve did not give a definitive answer (Fig. 17). Another possibility is that other cell surface proteins might be present at glial and axonal surfaces and act as integrin ligands. bind non-ECM ligands.  Vertebrate integrins have been shown to  For example, a neural cell adhesion molecule L1 has been  suggested to act as an integrin ligand to regulate Schwann cell myelination [302, 335]. It is unclear whether this non-ECM binding property is exclusive for vertebrate integrins, or it was inherited from ancient integrins and became more specialized to some vertebrate membranes.  The Drosophila homologues of these vertebrate integrin ligands and other  related proteins appear to be good candidates, which include, but not limited to Laminins, Thrombospondin, Tenascin major (homologue to Tenascin), SPARC, Kuzbanian (homologue to ADAM), Neuroglian (homologue to L1), FasII (homologue to NCAM), DSCAM, E-cadherin and N-cadherin.  To examine their potential as integrin ligands, we  need to trace these proteins with high specificity antibodies or to endogenously label proteins with protein traps and generation of direct knock-in alleles. 147  To test their  functions on glial development, we can conduct a RNAi screening of targeted genes in WG or neurons to look for WG phenotypes that mimic integrin RNAi.  4.3.2  Identification of integrin downstream signalling molecules  The cytoplasmic tail of integrin subunits are capable of interacting directly and indirectly with a large number of proteins.  A large integrin adhesome network map is  assembled that involves more than intracellular 150 components and 600 interactions [138]. The integrin signals can be transduced through physical ‘binding interactions’ to the actin cytoskeleton, or by activating or inhibitory ‘signalling interactions’ to different signalling pathways.  Although a lot of these integrin downstream partners have been  identified, only a subset of them are probably required in the Drosophila glia.  For  example, Talin, but not ILK and FAK, seems to be the direct effector downstream integrins in the two groups of glia that I examined.  To further investigate what  molecules are downstream of Talin, we can conduct a RNAi screen to target those Talin interacting proteins and relative genes in Drosophila. other enzymes,  For many kinases, GTPases and  we can also use repo-GAL4 to induce the expression of wild type,  hypomorphic or hypermorphic versions of these downstream candidate molecules.  The  caveats are that many proteins are commonly used in many glial cell types, including embryonic glia.  Robust GOF and LOF experiments might cause lethality before giving  any larval phenotypes.  Modified protocols might be required to activate the expression 148  in given glial subtypes using more specific GAL4 drivers, or in post-embryonic stages using temperature sensitive tub-GAL80.  4.3.3  Examination of glial-glial and glial-neuron interactions  I have shown that integrins are required cell autonomously by the PG and WG to form proper sheaths in the peripheral nerve.  However an unsolved question is how a  given type of glia affects other neighbouring glial types and neurons.  For example, the  WG send out numerous membrane processes to ensheath sensory and motor axons and these WG membranes are presumably acting as insulators to separate efferent and afferent axon fibres.  However action potentials along the motor axons seem to be  normal when most of the WG processes are lost in the absence of integrins (unpublished data).  An explanation is that these WG processes are dispensable for axonal firing.  Another possibility is that the SPG, which often send fine membrane processes to protrude into the internal regions between the axons and WG processes in control animals, now send more processes inside to compensate the loss of the WG ensheathment. To test whether the second possibility is true or not, we need to monitor SPG morphology while conducting the WG specific integrin RNAi.  However, the difficulty is that we  only have SPG specific GAL4 driver to label the SPG, which can’t be used with the WG specific GAL4 driver for the RNAi experiment.  In the future, we need to invest new  expression systems such as the Q system [336] or LexA/LexAop system [337] to label 149  SPG that are independent to the GAL4/UAS system, such as SPG-LexA/LexAop-GFP. Alternatively, the transmission electron microscope can be used to view SPG morphological changes in any given genotype. In the developing visual system, down-regulation of integrin or talin cause photoreceptor axon stalling phenotypes in the eye disc or the optic stalk.  It is still  unknown whether these integrin deficient glia make a physical or biochemical barrier to block axonal outgrowth.  A small targeted RNAi co-expression screen experiment was  conducted but failed to identify potential inhibitory cues in the glia (data not shown). The difficulty was that the penetrance of axonal stalling phenotype was sensitive to the RNAi efficiency as discussed in Chapter 3.  Co-expression of multiple UAS constructs,  including two UAS-RNAi lines and UAS-Dicer2, had significant dilution effects and made phenotypes volatile.  To avoid the dilution effect, we need to generate repo-QF  and QUAS-βPS-RNAi, or LexAop-βPS-RNAi (repo-lexA was available) to knock down integrins  in  glia  and  test  whether  they  can  give  similar  phenotype  as  repo-GAL4>UAS-βPS-RNAi. Then we can conduct a RNAi screen to down-regulate target genes using the repo-GAL4 and established UAS-RNAi collections. Overall in the future, by using the increasing RNAi and protein trap collections, as well as new direct genomic engineering method to generate knock-in alleles, I am able to confirm and identify potential integrin extracellular ligands and intracellular interacting proteins in glial cells.  By combining different binary expression systems, I will have the 150  power to manipulate two glial subtypes or glia and neurons at the same time and learn more about how glial cells use cell surface receptors and adhesion molecules, i.e. integrins, to orchestrate their development with their neighbouring cells during the formation of a complex nervous system.  151  Bibliography 1.  Virchow, R., Ueber das granulirte Aussehen der Wandungen der Gehirnventrikel. Allg Z Psychiat, 1846. 3: p. 9.  2.  Kettenmann, H., et al., Physiology of microglia. Physiol Rev, 2011. 91(2): p. 461-553.  3.  Jessen, K.R., Glial cells. Int J Biochem Cell Biol, 2004. 36(10): p. 1861-7.  4.  Kolb, H., Glial Cells of the Retina. 1995.  5.  Del Bigio, M.R., Ependymal cells: biology and pathology. Acta Neuropathol, 2010. 119(1): p. 55-73.  6.  Su, Z. and C. He, Olfactory ensheathing cells: biology in neural development and regeneration. Prog Neurobiol, 2010. 92(4): p. 517-32.  7.  Hanani, M., Satellite glial cells in sensory ganglia: from form to function. Brain Res Brain Res Rev, 2005. 48(3): p. 457-76.  8.  Ruhl, A., Glial cells in the gut. Neurogastroenterol Motil, 2005. 17(6): p. 777-90.  9.  Sretavan, D.W., Optic nerve formation, in Retianl development, E. Sernagor, Eglen, S., Harries,B. and Wong, R., Editor. 2006, Cambridge university press: Cambridge. p. 150-171.  10.  Butt, A.M., et al., Functions of optic nerve glia: axoglial signalling in physiology and pathology. Eye (Lond), 2004. 18(11): p. 1110-21.  11.  Raff, M.C., E.R. Abney, and R.H. Miller, Two glial cell lineages diverge prenatally in rat optic nerve. Dev Biol, 1984. 106(1): p. 53-60.  12.  Miller, R.H., et al., A quantitative immunohistochemical study of macroglial cell development in the rat optic nerve: in vivo evidence for two distinct astrocyte lineages. Dev Biol, 1985. 111(1): p. 35-41.  13.  Raff, M.C., R.H. Miller, and M. Noble, A glial progenitor cell that develops in vitro into an astrocyte or an oligodendrocyte depending on culture medium. Nature, 1983. 303(5916): p. 390-6.  14.  Fulton, B.P., J.F. Burne, and M.C. Raff, Visualization of O-2A progenitor cells in developing and adult rat optic nerve by quisqualate-stimulated cobalt uptake. J Neurosci, 1992. 12(12): p. 4816-33.  15.  Mi, H. and B.A. Barres, Purification and characterization of astrocyte precursor cells in the developing rat optic nerve. J Neurosci, 1999. 19(3): p. 1049-61.  16.  Burne, J.F. and M.C. Raff, Retinal ganglion cell axons drive the proliferation of astrocytes in the developing rodent optic nerve. Neuron, 1997. 18(2): p. 223-30.  17.  Burne, J.F., J.K. Staple, and M.C. Raff, Glial cells are increased proportionally in transgenic optic nerves with increased numbers of axons. J Neurosci, 1996. 16(6): p. 2064-73.  18.  Juurlink, B.H. and S. Fedoroff, Differentiation capabilities of mouse optic stalk in isolation of its immediate in vivo environment. Dev Biol, 1980. 78(1): p. 215-21.  19.  Mi, H., H. Haeberle, and B.A. Barres, Induction of astrocyte differentiation by endothelial cells. J Neurosci, 2001. 21(5): p. 1538-47. 152  20.  Dakubo, G.D., et al., Retinal ganglion cell-derived sonic hedgehog signaling is required for optic disc and stalk neuroepithelial cell development. Development, 2003. 130(13): p. 2967-80.  21.  Morcillo, J., et al., Proper patterning of the optic fissure requires the sequential activity of BMP7 and SHH. Development, 2006. 133(16): p. 3179-90.  22.  Wallace, V.A. and M.C. Raff, A role for Sonic hedgehog in axon-to-astrocyte signalling in the rodent optic nerve. Development, 1999. 126(13): p. 2901-9.  23.  Dakubo, G.D., et al., Control of glial precursor cell development in the mouse optic nerve by sonic hedgehog from retinal ganglion cells. Brain Res, 2008. 1228: p. 27-42.  24.  Zhang, Y. and B.A. Barres, Astrocyte heterogeneity: an underappreciated topic in neurobiology. Curr Opin Neurobiol, 2010. 20(5): p. 588-94.  25.  Butt, A.M. and B.R. Ransom, Morphology of astrocytes and oligodendrocytes during development in the intact rat optic nerve. J Comp Neurol, 1993. 338(1): p. 141-58.  26.  Wang, D.D. and A. Bordey, The astrocyte odyssey. Prog Neurobiol, 2008. 86(4): p. 342-67.  27.  Connors, B.W., et al., Activity-dependent K+ accumulation in the developing rat optic nerve. Science, 1982. 216(4552): p. 1341-3.  28.  Orkand, R.K., J.G. Nicholls, and S.W. Kuffler, Effect of nerve impulses on the membrane potential of glial cells in the central nervous system of amphibia. J Neurophysiol, 1966. 29(4): p. 788-806.  29.  Yamate, C.L. and B.R. Ransom, Effects of altered gliogenesis on activity-dependent K+ accumulation in the developing rat optic nerve. Brain Res, 1985. 353(2): p. 167-73.  30.  Ono, K., et al., Focal ventricular origin and migration of oligodendrocyte precursors into the chick optic nerve. Neuron, 1997. 19(2): p. 283-92.  31.  Sugimoto, Y., et al., Guidance of glial precursor cell migration by secreted cues in the developing optic nerve. Development, 2001. 128(17): p. 3321-30.  32.  Spassky, N., et al., Directional guidance of oligodendroglial migration by class 3 semaphorins and netrin-1. J Neurosci, 2002. 22(14): p. 5992-6004.  33.  Rowitch, D.H., Glial specification in the vertebrate neural tube. Nat Rev Neurosci, 2004. 5(5): p. 409-19.  34.  Rowitch, D.H. and A.R. Kriegstein, Developmental genetics of vertebrate glial-cell specification. Nature, 2010. 468(7321): p. 214-22.  35.  Garcion, E., A. Faissner, and C. ffrench-Constant, Knockout mice reveal a contribution of the extracellular matrix molecule tenascin-C to neural precursor proliferation and migration. Development, 2001. 128(13): p. 2485-96.  36.  Deiner, M.S., et al., Netrin-1 and DCC mediate axon guidance locally at the optic disc: loss of function leads to optic nerve hypoplasia. Neuron, 1997. 19(3): p. 575-89.  37.  Barres, B.A. and M.C. Raff, Proliferation of oligodendrocyte precursor cells depends on electrical activity in axons. Nature, 1993. 361(6409): p. 258-60.  38.  Barres, B.A. and M.C. Raff, Axonal control of oligodendrocyte development. J Cell Biol, 1999. 147(6): p. 1123-8.  39.  Fernandez, P.A., et al., Evidence that axon-derived neuregulin promotes oligodendrocyte survival in the developing rat optic nerve. Neuron, 2000. 28(1): p. 81-90. 153  40.  Wang, S., et al., Notch receptor activation inhibits oligodendrocyte differentiation. Neuron, 1998. 21(1): p. 63-75.  41.  Bansal, R., et al., Regulation of FGF receptors in the oligodendrocyte lineage. Mol Cell Neurosci, 1996. 7(4): p. 263-75.  42.  Qian, X., et al., FGF2 concentration regulates the generation of neurons and glia from multipotent cortical stem cells. Neuron, 1997. 18(1): p. 81-93.  43.  Barres, B.A., M.A. Lazar, and M.C. Raff, A novel role for thyroid hormone, glucocorticoids and retinoic acid in timing oligodendrocyte development. Development, 1994. 120(5): p. 1097-108.  44.  Payne, H.R. and V. Lemmon, Glial cells of the O-2A lineage bind preferentially to N-cadherin and develop distinct morphologies. Dev Biol, 1993. 159(2): p. 595-607.  45.  Barbin, G., et al., Axonal cell-adhesion molecule L1 in CNS myelination. Neuron Glia Biol, 2004. 1(1): p. 65-72.  46.  Charles, P., et al., Negative regulation of central nervous system myelination by polysialylated-neural cell adhesion molecule. Proc Natl Acad Sci U S A, 2000. 97(13): p. 7585-90.  47.  Barres, B.A., et al., Cell death and control of cell survival in the oligodendrocyte lineage. Cell, 1992. 70(1): p. 31-46.  48.  Barres, B.A., et al., Multiple extracellular signals are required for long-term oligodendrocyte survival. Development, 1993. 118(1): p. 283-95.  49.  Barres, B.A., et al., A crucial role for neurotrophin-3 in oligodendrocyte development. Nature, 1994. 367(6461): p. 371-5.  50.  Le Bras, B., et al., VEGF-C is a trophic factor for neural progenitors in the vertebrate embryonic brain. Nat Neurosci, 2006. 9(3): p. 340-8.  51.  Butt, A.M. and B.R. Ransom, Visualization of oligodendrocytes and astrocytes in the intact rat optic nerve by intracellular injection of lucifer yellow and horseradish peroxidase. Glia, 1989. 2(6): p. 470-5.  52.  Ransom, B.R., A.M. Butt, and J.A. Black, Ultrastructural identification of HRP-injected oligodendrocytes in the intact rat optic nerve. Glia, 1991. 4(1): p. 37-45.  53.  Kaplan, M.R., et al., Induction of sodium channel clustering by oligodendrocytes. Nature, 1997. 386(6626): p. 724-8.  54.  Kaplan, M.R., et al., Differential control of clustering of the sodium channels Na(v)1.2 and Na(v)1.6 at developing CNS nodes of Ranvier. Neuron, 2001. 30(1): p. 105-19.  55.  Rasband, M.N., et al., Dependence of nodal sodium channel clustering on paranodal axoglial contact in the developing CNS. J Neurosci, 1999. 19(17): p. 7516-28.  56.  Griffiths, I., et al., Current concepts of PLP and its role in the nervous system. Microsc Res Tech, 1998. 41(5): p. 344-58.  57.  Lappe-Siefke, C., et al., Disruption of Cnp1 uncouples oligodendroglial functions in axonal support and myelination. Nat Genet, 2003. 33(3): p. 366-74.  58.  Menichella, D.M., et al., Connexins are critical for normal myelination in the CNS. J Neurosci, 2003. 23(13): p. 5963-73.  59.  Jessen, K.R. and R. Mirsky, The origin and development of glial cells in peripheral nerves. Nat Rev 154  Neurosci, 2005. 6(9): p. 671-82. 60.  Murinson, B.B. and J.W. Griffin, C-fiber structure varies with location in peripheral nerve. J Neuropathol Exp Neurol, 2004. 63(3): p. 246-54.  61.  Nave, K.A. and J.L. Salzer, Axonal regulation of myelination by neuregulin 1. Curr Opin Neurobiol, 2006. 16(5): p. 492-500.  62.  Jessen, K.R. and R. Mirsky, Schwann cells and their precursors emerge as major regulators of nerve development. Trends Neurosci, 1999. 22(9): p. 402-10.  63.  Newbern, J. and C. Birchmeier, Nrg1/ErbB signaling networks in Schwann cell development and myelination. Semin Cell Dev Biol, 2010. 21(9): p. 922-8.  64.  Shah, N.M., et al., Glial growth factor restricts mammalian neural crest stem cells to a glial fate. Cell, 1994. 77(3): p. 349-60.  65.  Dong, Z., et al., Neu differentiation factor is a neuron-glia signal and regulates survival, proliferation, and maturation of rat Schwann cell precursors. Neuron, 1995. 15(3): p. 585-96.  66.  Wolpowitz, D., et al., Cysteine-rich domain isoforms of the neuregulin-1 gene are required for maintenance of peripheral synapses. Neuron, 2000. 25(1): p. 79-91.  67.  Grinspan, J.B., et al., Axonal interactions regulate Schwann cell apoptosis in developing peripheral nerve: neuregulin receptors and the role of neuregulins. J Neurosci, 1996. 16(19): p. 6107-18.  68.  Trachtenberg, J.T. and W.J. Thompson, Schwann cell apoptosis at developing neuromuscular junctions is regulated by glial growth factor. Nature, 1996. 379(6561): p. 174-7.  69.  Syroid, D.E., et al., Cell death in the Schwann cell lineage and its regulation by neuregulin. Proc Natl Acad Sci U S A, 1996. 93(17): p. 9229-34.  70.  Leimeroth, R., et al., Membrane-bound neuregulin1 type III actively promotes Schwann cell differentiation of multipotent Progenitor cells. Dev Biol, 2002. 246(2): p. 245-58.  71.  Meintanis, S., et al., The neuron-glia signal beta-neuregulin promotes Schwann cell motility via the MAPK pathway. Glia, 2001. 34(1): p. 39-51.  72.  Garratt, A.N., et al., A dual role of erbB2 in myelination and in expansion of the schwann cell precursor pool. J Cell Biol, 2000. 148(5): p. 1035-46.  73.  Michailov, G.V., et al., Axonal neuregulin-1 regulates myelin sheath thickness. Science, 2004. 304(5671): p. 700-3.  74.  Chen, S., et al., Neuregulin 1-erbB signaling is necessary for normal myelination and sensory function. J Neurosci, 2006. 26(12): p. 3079-86.  75.  Riethmacher, D., et al., Severe neuropathies in mice with targeted mutations in the ErbB3 receptor. Nature, 1997. 389(6652): p. 725-30.  76.  Britsch, S., et al., The transcription factor Sox10 is a key regulator of peripheral glial development. Genes Dev, 2001. 15(1): p. 66-78.  77.  Vermeren, M., et al., Integrity of developing spinal motor columns is regulated by neural crest derivatives at motor exit points. Neuron, 2003. 37(3): p. 403-15.  78.  Woldeyesus, M.T., et al., Peripheral nervous system defects in erbB2 mutants following genetic rescue of heart development. Genes Dev, 1999. 13(19): p. 2538-48.  79.  Morris, J.K., et al., Rescue of the cardiac defect in ErbB2 mutant mice reveals essential roles of 155  ErbB2 in peripheral nervous system development. Neuron, 1999. 23(2): p. 273-83. 80.  Parmantier, E., et al., Schwann cell-derived Desert hedgehog controls the development of peripheral nerve sheaths. Neuron, 1999. 23(4): p. 713-24.  81.  Azevedo, F.A., et al., Equal numbers of neuronal and nonneuronal cells make the human brain an isometrically scaled-up primate brain. J Comp Neurol, 2009. 513(5): p. 532-41.  82.  Pfrieger, F.W. and B.A. Barres, What the fly's glia tell the fly's brain. Cell, 1995. 83(5): p. 671-4.  83.  Freeman, M.R. and J. Doherty, Glial cell biology in Drosophila and vertebrates. Trends Neurosci, 2006. 29(2): p. 82-90.  84.  Hartenstein, V., Morphological diversity and development of glia in Drosophila. Glia, 2011. 59(9): p. 1237-52.  85.  Edenfeld, G., T. Stork, and C. Klambt, Neuron-glia interaction in the insect nervous system. Curr Opin Neurobiol, 2005. 15(1): p. 34-9.  86.  Pereanu, W., D. Shy, and V. Hartenstein, Morphogenesis and proliferation of the larval brain glia in Drosophila. Dev Biol, 2005. 283(1): p. 191-203.  87.  Klambt, C., J.R. Jacobs, and C.S. Goodman, The midline of the Drosophila central nervous system: a model for the genetic analysis of cell fate, cell migration, and growth cone guidance. Cell, 1991. 64(4): p. 801-15.  88.  Harris, R., L.M. Sabatelli, and M.A. Seeger, Guidance cues at the Drosophila CNS midline: identification and characterization of two Drosophila Netrin/UNC-6 homologs. Neuron, 1996. 17(2): p. 217-28.  89.  Mitchell, K.J., et al., Genetic analysis of Netrin genes in Drosophila: Netrins guide CNS commissural axons and peripheral motor axons. Neuron, 1996. 17(2): p. 203-15.  90.  Kolodziej, P.A., et al., frazzled encodes a Drosophila member of the DCC immunoglobulin subfamily and is required for CNS and motor axon guidance. Cell, 1996. 87(2): p. 197-204.  91.  Kidd, T., K.S. Bland, and C.S. Goodman, Slit is the midline repellent for the robo receptor in Drosophila. Cell, 1999. 96(6): p. 785-94.  92.  Sepp, K.J., J. Schulte, and V.J. Auld, Developmental dynamics of peripheral glia in Drosophila melanogaster. Glia, 2000. 30(2): p. 122-33.  93.  Sepp, K.J., J. Schulte, and V.J. Auld, Peripheral glia direct axon guidance across the CNS/PNS transition zone. Dev Biol, 2001. 238(1): p. 47-63.  94.  Poeck, B., et al., Glial cells mediate target layer selection of retinal axons in the developing visual system of Drosophila. Neuron, 2001. 29(1): p. 99-113.  95.  Tayler, T.D., M.B. Robichaux, and P.A. Garrity, Compartmentalization of visual centers in the Drosophila brain requires Slit and Robo proteins. Development, 2004. 131(23): p. 5935-45.  96.  Watts, R.J., et al., Glia engulf degenerating axons during developmental axon pruning. Curr Biol, 2004. 14(8): p. 678-84.  97.  Awasaki, T., et al., Glia instruct developmental neuronal remodeling through TGF-beta signaling. Nat Neurosci, 2011. 14(7): p. 821-3.  98.  Seal, R.P., et al., Identification and characterization of a cDNA encoding a neuronal glutamate transporter from Drosophila melanogaster. Receptors Channels, 1998. 6(1): p. 51-64. 156  99.  Besson, M.T., L. Soustelle, and S. Birman, Identification and structural characterization of two genes encoding glutamate transporter homologues differently expressed in the nervous system of Drosophila melanogaster. FEBS Lett, 1999. 443(2): p. 97-104.  100.  Kawano, T., et al., Cloning and characterization of a Drosophila melanogaster cDNA encoding a glutamate transporter. Biosci Biotechnol Biochem, 1999. 63(11): p. 2042-4.  101.  Rival, T., et al., Decreasing glutamate buffering capacity triggers oxidative stress and neuropil degeneration in the Drosophila brain. Curr Biol, 2004. 14(7): p. 599-605.  102.  De Pinto, V., et al., Purification of the glutamine synthetase II isozyme of Drosophila melanogaster and structural and functional comparison of glutamine synthetases I and II. Biochem Genet, 1987. 25(11-12): p. 821-36.  103.  Caizzi, R., et al., Homologous nuclear genes encode cytoplasmic and mitochondrial glutamine synthetase in Drosophila melanogaster. J Mol Biol, 1990. 212(1): p. 17-26.  104.  Caggese, C., et al., Mutations in the glutamine synthetase I (gsI) gene produce embryo-lethal female sterility in Drosophila melanogaster. Dev Genet, 1992. 13(5): p. 359-66.  105.  Caggese, C., et al., Genetic, molecular and developmental analysis of the glutamine synthetase isozymes of Drosophila melanogaster. Genetica, 1994. 94(2-3): p. 275-81.  106.  Stacey, S.M., et al., Drosophila glial glutamate transporter Eaat1 is regulated by fringe-mediated notch signaling and is essential for larval locomotion. J Neurosci, 2010. 30(43): p. 14446-57.  107.  Auld, V.J., et al., Gliotactin, a novel transmembrane protein on peripheral glia, is required to form the blood-nerve barrier in Drosophila. Cell, 1995. 81(5): p. 757-67.  108.  Baumgartner, S., et al., A Drosophila neurexin is required for septate junction and blood-nerve barrier formation and function. Cell, 1996. 87(6): p. 1059-68.  109.  Stork, T., et al., Organization and function of the blood-brain barrier in Drosophila. J Neurosci, 2008. 28(3): p. 587-97.  110.  MacDonald, J.M., et al., The Drosophila cell corpse engulfment receptor Draper mediates glial clearance of severed axons. Neuron, 2006. 50(6): p. 869-81.  111.  Jones, B.W., et al., glial cells missing: a genetic switch that controls glial versus neuronal fate. Cell, 1995. 82(6): p. 1013-23.  112.  Hosoya, T., et al., glial cells missing: a binary switch between neuronal and glial determination in Drosophila. Cell, 1995. 82(6): p. 1025-36.  113.  Vincent, S., J.L. Vonesch, and A. Giangrande, Glide directs glial fate commitment and cell fate switch between neurones and glia. Development, 1996. 122(1): p. 131-9.  114.  Akiyama-Oda, Y., T. Hosoya, and Y. Hotta, Alteration of cell fate by ectopic expression of Drosophila glial cells missing in non-neural cells. Dev Genes Evol, 1998. 208(10): p. 578-85.  115.  Schmidt, H., et al., The embryonic central nervous system lineages of Drosophila melanogaster. II. Neuroblast lineages derived from the dorsal part of the neuroectoderm. Dev Biol, 1997. 189(2): p. 186-204.  116.  Edwards, J.S., L.S. Swales, and M. Bate, The differentiation between neuroglia and connective tissue sheath in insect ganglia revisited: the neural lamella and perineurial sheath cells are absent in a mesodermless mutant of Drosophila. J Comp Neurol, 1993. 333(2): p. 301-8. 157  117.  Unhavaithaya, Y. and T.L. Orr-Weaver, Polyploidization of glia in neural development links tissue growth to blood-brain barrier integrity. Genes Dev, 2012. 26(1): p. 31-6.  118.  Leiserson, W.M., E.W. Harkins, and H. Keshishian, Fray, a Drosophila serine/threonine kinase homologous to mammalian PASK, is required for axonal ensheathment. Neuron, 2000. 28(3): p. 793-806.  119.  Banerjee, S., et al., Axonal ensheathment and septate junction formation in the peripheral nervous system of Drosophila. J Neurosci, 2006. 26(12): p. 3319-29.  120.  Faivre-Sarrailh, C., et al., Drosophila contactin, a homolog of vertebrate contactin, is required for septate junction organization and paracellular barrier function. Development, 2004. 131(20): p. 4931-42.  121.  Olofsson, B. and D.T. Page, Condensation of the central nervous system in embryonic Drosophila is inhibited by blocking hemocyte migration or neural activity. Dev Biol, 2005. 279(1): p. 233-43.  122.  Pappu, K.S. and G. Mardon, Genetic control of retinal specification and determination in Drosophila. Int J Dev Biol, 2004. 48(8-9): p. 913-24.  123.  Tayler, T.D. and P.A. Garrity, Axon targeting in the Drosophila visual system. Curr Opin Neurobiol, 2003. 13(1): p. 90-5.  124.  Rangarajan, R., Q. Gong, and U. Gaul, Migration and function of glia in the developing Drosophila eye. Development, 1999. 126(15): p. 3285-92.  125.  Silies, M., et al., The eye imaginal disc as a model to study the coordination of neuronal and glial development. Fly (Austin), 2010. 4(1): p. 71-9.  126.  Silies, M., et al., Glial cell migration in the eye disc. J Neurosci, 2007. 27(48): p. 13130-9.  127.  Choi, K.W. and S. Benzer, Migration of glia along photoreceptor axons in the developing Drosophila eye. Neuron, 1994. 12(2): p. 423-31.  128.  Hummel, T., et al., Temporal control of glial cell migration in the Drosophila eye requires gilgamesh, hedgehog, and eye specification genes. Neuron, 2002. 33(2): p. 193-203.  129.  Rangarajan, R., H. Courvoisier, and U. Gaul, Dpp and Hedgehog mediate neuron-glia interactions in Drosophila eye development by promoting the proliferation and motility of subretinal glia. Mech Dev, 2001. 108(1-2): p. 93-103.  130.  Franzdottir, S.R., et al., Switch in FGF signalling initiates glial differentiation in the Drosophila eye. Nature, 2009. 460(7256): p. 758-61.  131.  Hynes, R.O. and Q. Zhao, The evolution of cell adhesion. J Cell Biol, 2000. 150(2): p. F89-96.  132.  Danen, E.H.J., Integrins: an overview of structural and functional aspects, in Integrins and development, E.H.J. Danen, Editor. 2006, Landes Bioscience. p. 1-15.  133.  Xiong, J.P., et al., Crystal structure of the extracellular segment of integrin alpha Vbeta3. Science, 2001. 294(5541): p. 339-45.  134.  Xiong, J.P., et al., Crystal structure of the extracellular segment of integrin alpha Vbeta3 in complex with an Arg-Gly-Asp ligand. Science, 2002. 296(5565): p. 151-5.  135.  Xiao, T., et al., Structural basis for allostery in integrins and binding to fibrinogen-mimetic therapeutics. Nature, 2004. 432(7013): p. 59-67.  136.  Arnaout, M.A., B. Mahalingam, and J.P. Xiong, Integrin structure, allostery, and bidirectional 158  signaling. Annu Rev Cell Dev Biol, 2005. 21: p. 381-410. 137.  Shimaoka, M., J. Takagi, and T.A. Springer, Conformational regulation of integrin structure and function. Annu Rev Biophys Biomol Struct, 2002. 31: p. 485-516.  138.  Zaidel-Bar, R., et al., Functional atlas of the integrin adhesome. Nat Cell Biol, 2007. 9(8): p. 858-67.  139.  Legate, K.R., et al., ILK, PINCH and parvin: the tIPP of integrin signalling. Nat Rev Mol Cell Biol, 2006. 7(1): p. 20-31.  140.  Brower, D.L., et al., Molecular evolution of integrins: genes encoding integrin beta subunits from a coral and a sponge. Proc Natl Acad Sci U S A, 1997. 94(17): p. 9182-7.  141.  Pancer, Z., et al., On the origin of Metazoan adhesion receptors: cloning of integrin alpha subunit from the sponge Geodia cydonium. Mol Biol Evol, 1997. 14(4): p. 391-8.  142.  Wimmer, W., et al., Origin of the integrin-mediated signal transduction. Functional studies with cell cultures from the sponge Suberites domuncula. Eur J Biochem, 1999. 260(1): p. 156-65.  143.  Brown, N.H., Cell-cell adhesion via the ECM: integrin genetics in fly and worm. Matrix Biol, 2000. 19(3): p. 191-201.  144.  DiPersio, C.M., et al., Mouse keratinocytes immortalized with large T antigen acquire alpha3beta1 integrin-dependent secretion of MMP-9/gelatinase B. J Cell Sci, 2000. 113 ( Pt 16): p. 2909-21.  145.  Evans, R., et al., Integrins in immunity. J Cell Sci, 2009. 122(Pt 2): p. 215-25.  146.  Pribila, J.T., et al., Integrins and T cell-mediated immunity. Annu Rev Immunol, 2004. 22: p. 157-80.  147.  Hotchin, N.A., A. Gandarillas, and F.M. Watt, Regulation of cell surface beta 1 integrin levels during keratinocyte terminal differentiation. J Cell Biol, 1995. 128(6): p. 1209-19.  148.  Gohel, A.R., A.R. Hand, and G.A. Gronowicz, Immunogold localization of beta 1-integrin in bone: effect of glucocorticoids and insulin-like growth factor I on integrins and osteocyte formation. J Histochem Cytochem, 1995. 43(11): p. 1085-96.  149.  Durr, J., et al., Localization of beta 1-integrins in human cartilage and their role in chondrocyte adhesion to collagen and fibronectin. Exp Cell Res, 1993. 207(2): p. 235-44.  150.  van der Flier, A., et al., A novel beta 1 integrin isoform produced by alternative splicing: unique expression in cardiac and skeletal muscle. FEBS Lett, 1995. 369(2-3): p. 340-4.  151.  Zhidkova, N.I., A.M. Belkin, and R. Mayne, Novel isoform of beta 1 integrin expressed in skeletal and cardiac muscle. Biochem Biophys Res Commun, 1995. 214(1): p. 279-85.  152.  Albelda, S.M., et al., Identification and characterization of cell-substratum adhesion receptors on cultured human endothelial cells. J Clin Invest, 1989. 83(6): p. 1992-2002.  153.  Milner, R. and I.L. Campbell, The integrin family of cell adhesion molecules has multiple functions within the CNS. J Neurosci Res, 2002. 69(3): p. 286-91.  154.  Sobeih, M.M. and G. Corfas, Extracellular factors that regulate neuronal migration in the central nervous system. Int J Dev Neurosci, 2002. 20(3-5): p. 349-57.  155.  Berti, C., et al., Role of integrins in peripheral nerves and hereditary neuropathies. Neuromolecular Med, 2006. 8(1-2): p. 191-204.  156.  Shimaoka, M. and T.A. Springer, Therapeutic antagonists and conformational regulation of integrin function. Nat Rev Drug Discov, 2003. 2(9): p. 703-16. 159  157.  Cardwell, M.C. and L.H. Rome, Evidence that an RGD-dependent receptor mediates the binding of oligodendrocytes to a novel ligand in a glial-derived matrix. J Cell Biol, 1988. 107(4): p. 1541-9.  158.  Cardwell, M.C. and L.H. Rome, RGD-containing peptides inhibit the synthesis of myelin-like membrane by cultured oligodendrocytes. J Cell Biol, 1988. 107(4): p. 1551-9.  159.  Malek-Hedayat, S. and L.H. Rome, Expression of a beta 1-related integrin by oligodendroglia in primary culture: evidence for a functional role in myelination. J Cell Biol, 1994. 124(6): p. 1039-46.  160.  Milner, R. and C. Ffrench-Constant, A developmental analysis of oligodendroglial integrins in primary cells: changes in alpha v-associated beta subunits during differentiation. Development, 1994. 120(12): p. 3497-506.  161.  Milner, R., et al., Expression of alpha vbeta3 and alpha vbeta8 integrins during oligodendrocyte precursor differentiation in the presence and absence of axons. Glia, 1997. 21(4): p. 350-60.  162.  Milner, R., et al., A role in migration for the alpha V beta 1 integrin expressed on oligodendrocyte precursors. J Neurosci, 1996. 16(22): p. 7240-52.  163.  Buttery, P.C. and C. ffrench-Constant, Laminin-2/integrin interactions enhance myelin membrane formation by oligodendrocytes. Mol Cell Neurosci, 1999. 14(3): p. 199-212.  164.  Corley, S.M., et al., Astrocytes attenuate oligodendrocyte death in vitro through an alpha(6) integrin-laminin-dependent mechanism. Glia, 2001. 36(3): p. 281-94.  165.  Frost, E.E., et al., Integrins mediate a neuronal survival signal for oligodendrocytes. Curr Biol, 1999. 9(21): p. 1251-4.  166.  Relvas, J.B., et al., Expression of dominant-negative and chimeric subunits reveals an essential role for beta1 integrin during myelination. Curr Biol, 2001. 11(13): p. 1039-43.  167.  Tiwari-Woodruff, S.K., et al., OSP/claudin-11 forms a complex with a novel member of the tetraspanin super family and beta1 integrin and regulates proliferation and migration of oligodendrocytes. J Cell Biol, 2001. 153(2): p. 295-305.  168.  Benninger, Y., et al., Beta1-integrin signaling mediates premyelinating oligodendrocyte survival but is not required for CNS myelination and remyelination. J Neurosci, 2006. 26(29): p. 7665-73.  169.  Lee, K.K., et al., Dominant-negative beta1 integrin mice have region-specific myelin defects accompanied by alterations in MAPK activity. Glia, 2006. 53(8): p. 836-44.  170.  Barros, C.S., et al., Beta1 integrins are required for normal CNS myelination and promote AKT-dependent myelin outgrowth. Development, 2009. 136(16): p. 2717-24.  171.  Camara, J., et al., Integrin-mediated axoglial interactions initiate myelination in the central nervous system. J Cell Biol, 2009. 185(4): p. 699-712.  172.  Chun, S.J., et al., Integrin-linked kinase is required for laminin-2-induced oligodendrocyte cell spreading and CNS myelination. J Cell Biol, 2003. 163(2): p. 397-408.  173.  Colognato, H., et al., CNS integrins switch growth factor signalling to promote target-dependent survival. Nat Cell Biol, 2002. 4(11): p. 833-41.  174.  Jones, K.J., et al., The expanding phenotype of laminin alpha2 chain (merosin) abnormalities: case series and review. J Med Genet, 2001. 38(10): p. 649-57.  175.  Baron, W., S.J. Shattil, and C. ffrench-Constant, The oligodendrocyte precursor mitogen PDGF stimulates proliferation by activation of alpha(v)beta3 integrins. EMBO J, 2002. 21(8): p. 1957-66. 160  176.  Colognato, H., et al., Integrins direct Src family kinases to regulate distinct phases of oligodendrocyte development. J Cell Biol, 2004. 167(2): p. 365-75.  177.  Bradley, W.G. and M. Jenkison, Abnormalities of peripheral nerves in murine muscular dystrophy. J Neurol Sci, 1973. 18(2): p. 227-47.  178.  Stirling, C.A., Abnormalities in Schwann cell sheaths in spinal nerve roots of dystrophic mice. J Anat, 1975. 119(Pt 1): p. 169-80.  179.  Xu, H., et al., Murine muscular dystrophy caused by a mutation in the laminin alpha 2 (Lama2) gene. Nat Genet, 1994. 8(3): p. 297-302.  180.  Helbling-Leclerc, A., et al., Mutations in the laminin alpha 2-chain gene (LAMA2) cause merosin-deficient congenital muscular dystrophy. Nat Genet, 1995. 11(2): p. 216-8.  181.  Shorer, Z., et al., Demyelinating peripheral neuropathy in merosin-deficient congenital muscular dystrophy. J Child Neurol, 1995. 10(6): p. 472-5.  182.  Sunada, Y., et al., Identification of a novel mutant transcript of laminin alpha 2 chain gene responsible for muscular dystrophy and dysmyelination in dy2J mice. Hum Mol Genet, 1995. 4(6): p. 1055-61.  183.  Lentz, S.I., et al., Distribution of the ten known laminin chains in the pathways and targets of developing sensory axons. J Comp Neurol, 1997. 378(4): p. 547-61.  184.  Wallquist, W., et al., Laminin chains in rat and human peripheral nerve: distribution and regulation during development and after axonal injury. J Comp Neurol, 2002. 454(3): p. 284-93.  185.  Previtali, S.C., et al., Expression of laminin receptors in schwann cell differentiation: evidence for distinct roles. J Neurosci, 2003. 23(13): p. 5520-30.  186.  Desban, N. and J.L. Duband, Avian neural crest cell migration on laminin: interaction of the alpha1beta1 integrin with distinct laminin-1 domains mediates different adhesive responses. J Cell Sci, 1997. 110 ( Pt 21): p. 2729-44.  187.  Testaz, S., M. Delannet, and J. Duband, Adhesion and migration of avian neural crest cells on fibronectin require the cooperating activities of multiple integrins of the (beta)1 and (beta)3 families. J Cell Sci, 1999. 112 ( Pt 24): p. 4715-28.  188.  Lallier, T. and M. Bronner-Fraser, Inhibition of neural crest cell attachment by integrin antisense oligonucleotides. Science, 1993. 259(5095): p. 692-5.  189.  Kil, S.H., T. Lallier, and M. Bronner-Fraser, Inhibition of cranial neural crest adhesion in vitro and migration in vivo using integrin antisense oligonucleotides. Dev Biol, 1996. 179(1): p. 91-101.  190.  Perris, R. and D. Perissinotto, Role of the extracellular matrix during neural crest cell migration. Mech Dev, 2000. 95(1-2): p. 3-21.  191.  Pietri, T., et al., Conditional beta1-integrin gene deletion in neural crest cells causes severe developmental alterations of the peripheral nervous system. Development, 2004. 131(16): p. 3871-83.  192.  Haack, H. and R.O. Hynes, Integrin receptors are required for cell survival and proliferation during development of the peripheral glial lineage. Dev Biol, 2001. 233(1): p. 38-55.  193.  Feltri, M.L., et al., Conditional disruption of beta 1 integrin in Schwann cells impedes interactions with axons. J Cell Biol, 2002. 156(1): p. 199-209. 161  194.  Pereira, J.A., et al., Integrin-linked kinase is required for radial sorting of axons and Schwann cell remyelination in the peripheral nervous system. J Cell Biol, 2009. 185(1): p. 147-61.  195.  Nodari, A., et al., Beta1 integrin activates Rac1 in Schwann cells to generate radial lamellae during axonal sorting and myelination. J Cell Biol, 2007. 177(6): p. 1063-75.  196.  Milner, R., et al., Division of labor of Schwann cell integrins during migration on peripheral nerve extracellular matrix ligands. Dev Biol, 1997. 185(2): p. 215-28.  197.  Yu, W.M., et al., Disruption of laminin in the peripheral nervous system impedes nonmyelinating Schwann cell development and impairs nociceptive sensory function. Glia, 2009. 57(8): p. 850-9.  198.  Brower, D.L., et al., Related cell-surface antigens expressed with positional specificity in Drosophila imaginal discs. Proc Natl Acad Sci U S A, 1984. 81(23): p. 7485-9.  199.  Leptin, M., R. Aebersold, and M. Wilcox, Drosophila position-specific antigens resemble the vertebrate fibronectin-receptor family. EMBO J, 1987. 6(4): p. 1037-43.  200.  Takada, Y., X. Ye, and S. Simon, The integrins. Genome Biol, 2007. 8(5): p. 215.  201.  Bunch, T.A. and D.L. Brower, Drosophila PS2 integrin mediates RGD-dependent cell-matrix interactions. Development, 1992. 116(1): p. 239-47.  202.  Gotwals, P.J., et al., Drosophila PS1 integrin is a laminin receptor and differs in ligand specificity from PS2. Proc Natl Acad Sci U S A, 1994. 91(24): p. 11447-51.  203.  Stark, K.A., et al., A novel alpha integrin subunit associates with betaPS and functions in tissue morphogenesis and movement during Drosophila development. Development, 1997. 124(22): p. 4583-94.  204.  Schock, F. and N. Perrimon, Retraction of the Drosophila germ band requires cell-matrix interaction. Genes Dev, 2003. 17(5): p. 597-602.  205.  Leptin, M., et al., The function of PS integrins during Drosophila embryogenesis. Cell, 1989. 56(3): p. 401-8.  206.  Yee, G.H. and R.O. Hynes, A novel, tissue-specific integrin subunit, beta nu, expressed in the midgut of Drosophila melanogaster. Development, 1993. 118(3): p. 845-58.  207.  Brabant, M.C., et al., Distinct spatial and temporal functions for PS integrins during Drosophila wing morphogenesis. Development, 1996. 122(10): p. 3307-17.  208.  Bradley, P.L., et al., Posterior migration of the salivary gland requires an intact visceral mesoderm and integrin function. Dev Biol, 2003. 257(2): p. 249-62.  209.  Martin-Bermudo, M.D., I. Alvarez-Garcia, and N.H. Brown, Migration of the Drosophila primordial midgut cells requires coordination of diverse PS integrin functions. Development, 1999. 126(22): p. 5161-9.  210.  Hoang, B. and A. Chiba, Genetic analysis on the role of integrin during axon guidance in Drosophila. J Neurosci, 1998. 18(19): p. 7847-55.  211.  Grotewiel, M.S., et al., Integrin-mediated short-term memory in Drosophila. Nature, 1998. 391(6666): p. 455-60.  212.  Tanentzapf, G., et al., Integrin-dependent anchoring of a stem-cell niche. Nat Cell Biol, 2007. 9(12): p. 1413-8.  213.  O'Reilly, A.M., H.H. Lee, and M.A. Simon, Integrins control the positioning and proliferation of 162  follicle stem cells in the Drosophila ovary. J Cell Biol, 2008. 182(4): p. 801-15. 214.  Fristrom, D., M. Wilcox, and J. Fristrom, The distribution of PS integrins, laminin A and F-actin during key stages in Drosophila wing development. Development, 1993. 117(2): p. 509-23.  215.  Brower, D.L., et al., Nonequivalent requirements for PS1 and PS2 integrin at cell attachments in Drosophila: genetic analysis of the alpha PS1 integrin subunit. Development, 1995. 121(5): p. 1311-20.  216.  Brower, D.L. and S.M. Jaffe, Requirement for integrins during Drosophila wing development. Nature, 1989. 342(6247): p. 285-7.  217.  Wilcox, M., A. DiAntonio, and M. Leptin, The function of PS integrins in Drosophila wing morphogenesis. Development, 1989. 107(4): p. 891-7.  218.  Brabant, M.C. and D.L. Brower, PS2 integrin requirements in Drosophila embryo and wing morphogenesis. Dev Biol, 1993. 157(1): p. 49-59.  219.  Bate, M. and E. Rushton, Myogenesis and muscle patterning in Drosophila. C R Acad Sci III, 1993. 316(9): p. 1047-61.  220.  Bogaert, T., N. Brown, and M. Wilcox, The Drosophila PS2 antigen is an invertebrate integrin that, like the fibronectin receptor, becomes localized to muscle attachments. Cell, 1987. 51(6): p. 929-40.  221.  Wright, T.R., The phenogenetics of the embryonic mutant, lethal myospheroid, in Drosophila melanogaster. J Exp Zool, 1960. 143: p. 77-99.  222.  Newman, S.M., Jr. and T.R. Wright, A histological and ultrastructural analysis of developmental defects produced by the mutation, lethal(1)myospheroid, in Drosophila melanogaster. Dev Biol, 1981. 86(2): p. 393-402.  223.  Brown, N.H., Null mutations in the alpha PS2 and beta PS integrin subunit genes have distinct phenotypes. Development, 1994. 120(5): p. 1221-31.  224.  Yuan, L., et al., Analysis of integrin turnover in fly myotendinous junctions. J Cell Sci, 2010. 123(Pt 6): p. 939-46.  225.  Sparrow, J.C. and F. Schock, The initial steps of myofibril assembly: integrins pave the way. Nat Rev Mol Cell Biol, 2009. 10(4): p. 293-8.  226.  Perkins, A.D., et al., Integrin-mediated adhesion maintains sarcomeric integrity. Dev Biol, 2010. 338(1): p. 15-27.  227.  Reuter, R., B. Grunewald, and M. Leptin, A role for the mesoderm in endodermal migration and morphogenesis in Drosophila. Development, 1993. 119(4): p. 1135-45.  228.  Tepass, U. and V. Hartenstein, Epithelium formation in the Drosophila midgut depends on the interaction of endoderm and mesoderm. Development, 1994. 120(3): p. 579-90.  229.  Roote, C.E. and S. Zusman, Functions for PS integrins in tissue adhesion, migration, and shape changes during early embryonic development in Drosophila. Dev Biol, 1995. 169(1): p. 322-36.  230.  Devenport, D. and N.H. Brown, Morphogenesis in the absence of integrins: mutation of both Drosophila beta subunits prevents midgut migration. Development, 2004. 131(21): p. 5405-15.  231.  Manning, G.a.K., M.A., Development of the Drosophila tracheal system., in The development of Drosophila melanogaster., M.B.a.A. Martínez-Arias, Editor. 1993, Cold Spring Harbor Laboratory 163  Press: Cold Spring Harbor. p. 609-685. 232.  Boube, M., et al., Specific tracheal migration is mediated by complementary expression of cell surface proteins. Genes Dev, 2001. 15(12): p. 1554-62.  233.  Aigouy, B., et al., Time-lapse and cell ablation reveal the role of cell interactions in fly glia migration and proliferation. Development, 2004. 131(20): p. 5127-38.  234.  Fogerty, F.J., et al., Tiggrin, a novel Drosophila extracellular matrix protein that functions as a ligand for Drosophila alpha PS2 beta PS integrins. Development, 1994. 120(7): p. 1747-58.  235.  Ilic, D., et al., Reduced cell motility and enhanced focal adhesion contact formation in cells from FAK-deficient mice. Nature, 1995. 377(6549): p. 539-44.  236.  Grabbe, C., et al., Focal adhesion kinase is not required for integrin function or viability in Drosophila. Development, 2004. 131(23): p. 5795-805.  237.  Aumailley, M., et al., A simplified laminin nomenclature. Matrix Biol, 2005. 24(5): p. 326-32.  238.  Miner, J.H. and P.D. Yurchenco, Laminin functions in tissue morphogenesis. Annu Rev Cell Dev Biol, 2004. 20: p. 255-84.  239.  Kusche-Gullberg, M., et al., Laminin A chain: expression during Drosophila development and genomic sequence. EMBO J, 1992. 11(12): p. 4519-27.  240.  Martin, D., et al., wing blister, a new Drosophila laminin alpha chain required for cell adhesion and migration during embryonic and imaginal development. J Cell Biol, 1999. 145(1): p. 191-201.  241.  Montell, D.J. and C.S. Goodman, Drosophila substrate adhesion molecule: sequence of laminin B1 chain reveals domains of homology with mouse. Cell, 1988. 53(3): p. 463-73.  242.  Montell, D.J. and C.S. Goodman, Drosophila laminin: sequence of B2 subunit and expression of all three subunits during embryogenesis. J Cell Biol, 1989. 109(5): p. 2441-53.  243.  Graner, M.W., et al., Splice variants of the Drosophila PS2 integrins differentially interact with RGD-containing fragments of the extracellular proteins tiggrin, ten-m, and D-laminin 2. J Biol Chem, 1998. 273(29): p. 18235-41.  244.  Fessler, L.I., et al., Drosophila laminin: characterization and localization. J Cell Biol, 1987. 105(5): p. 2383-91.  245.  Henchcliffe, C., et al., Genetic analysis of laminin A reveals diverse functions during morphogenesis in Drosophila. Development, 1993. 118(2): p. 325-37.  246.  Yarnitzky, T. and T. Volk, Laminin is required for heart, somatic muscles, and gut development in the Drosophila embryo. Dev Biol, 1995. 169(2): p. 609-18.  247.  Rees, D.J., et al., Sequence and domain structure of talin. Nature, 1990. 347(6294): p. 685-9.  248.  Zhang, X., et al., Talin depletion reveals independence of initial cell spreading from integrin activation and traction. Nat Cell Biol, 2008. 10(9): p. 1062-8.  249.  Senetar, M.A. and R.O. McCann, Gene duplication and functional divergence during evolution of the cytoskeletal linker protein talin. Gene, 2005. 362: p. 141-52.  250.  Brown, N.H., et al., Talin is essential for integrin function in Drosophila. Dev Cell, 2002. 3(4): p. 569-79.  251.  Calderwood, D.A. and M.H. Ginsberg, Talin forges the links between integrins and actin. Nat Cell Biol, 2003. 5(8): p. 694-7. 164  252.  Critchley, D.R., Biochemical and structural properties of the integrin-associated cytoskeletal protein talin. Annu Rev Biophys, 2009. 38: p. 235-54.  253.  Critchley, D.R., Cytoskeletal proteins talin and vinculin in integrin-mediated adhesion. Biochem Soc Trans, 2004. 32(Pt 5): p. 831-6.  254.  Monkley, S.J., et al., Disruption of the talin gene arrests mouse development at the gastrulation stage. Dev Dyn, 2000. 219(4): p. 560-74.  255.  Nieswandt, B., et al., Loss of talin1 in platelets abrogates integrin activation, platelet aggregation, and thrombus formation in vitro and in vivo. J Exp Med, 2007. 204(13): p. 3113-8.  256.  Petrich, B.G., et al., Talin is required for integrin-mediated platelet function in hemostasis and thrombosis. J Exp Med, 2007. 204(13): p. 3103-11.  257.  Conti, F.J., et al., Progressive myopathy and defects in the maintenance of myotendinous junctions in mice that lack talin 1 in skeletal muscle. Development, 2008. 135(11): p. 2043-53.  258.  Chen, N.T. and S.H. Lo, The N-terminal half of talin2 is sufficient for mouse development and survival. Biochem Biophys Res Commun, 2005. 337(2): p. 670-6.  259.  Levi, B.P., A.S. Ghabrial, and M.A. Krasnow, Drosophila talin and integrin genes are required for maintenance of tracheal terminal branches and luminal organization. Development, 2006. 133(12): p. 2383-93.  260.  Tanentzapf, G. and N.H. Brown, An interaction between integrin and the talin FERM domain mediates integrin activation but not linkage to the cytoskeleton. Nat Cell Biol, 2006. 8(6): p. 601-6.  261.  Becam, I.E., et al., Integrin-independent repression of cadherin transcription by talin during axis formation in Drosophila. Nat Cell Biol, 2005. 7(5): p. 510-6.  262.  Hannigan, G.E., et al., Regulation of cell adhesion and anchorage-dependent growth by a new beta 1-integrin-linked protein kinase. Nature, 1996. 379(6560): p. 91-6.  263.  Nikolopoulos, S.N. and C.E. Turner, Integrin-linked kinase (ILK) binding to paxillin LD1 motif regulates ILK localization to focal adhesions. J Biol Chem, 2001. 276(26): p. 23499-505.  264.  Sakai, T., et al., Integrin-linked kinase (ILK) is required for polarizing the epiblast, cell adhesion, and controlling actin accumulation. Genes Dev, 2003. 17(7): p. 926-40.  265.  Stephens, L.E., et al., Deletion of beta 1 integrins in mice results in inner cell mass failure and peri-implantation lethality. Genes Dev, 1995. 9(15): p. 1883-95.  266.  Fassler, R. and M. Meyer, Consequences of lack of beta 1 integrin gene expression in mice. Genes Dev, 1995. 9(15): p. 1896-908.  267.  Zervas, C.G., S.L. Gregory, and N.H. Brown, Drosophila integrin-linked kinase is required at sites of integrin adhesion to link the cytoskeleton to the plasma membrane. J Cell Biol, 2001. 152(5): p. 1007-18.  268.  Olsson, Y., Microenvironment of the peripheral nervous system under normal and pathological conditions. Crit Rev Neurobiol, 1990. 5(3): p. 265-311.  269.  Lavery, W., et al., Phosphatidylinositol 3-kinase and Akt nonautonomously promote perineurial glial growth in Drosophila peripheral nerves. J Neurosci, 2007. 27(2): p. 279-88.  270.  Bhat, M.A., Molecular organization of axo-glial junctions. Curr Opin Neurobiol, 2003. 13(5): p. 552-9. 165  271.  Parker, R.J. and V.J. Auld, Roles of glia in the Drosophila nervous system. Semin Cell Dev Biol, 2006. 17(1): p. 66-77.  272.  Chen, Z.L. and S. Strickland, Laminin gamma1 is critical for Schwann cell differentiation, axon myelination, and regeneration in the peripheral nerve. J Cell Biol, 2003. 163(4): p. 889-99.  273.  Wallquist, W., et al., Impeded interaction between Schwann cells and axons in the absence of laminin alpha4. J Neurosci, 2005. 25(14): p. 3692-700.  274.  Yang, D., et al., Coordinate control of axon defasciculation and myelination by laminin-2 and -8. J Cell Biol, 2005. 168(4): p. 655-66.  275.  Yu, W.M., et al., Schwann cell-specific ablation of laminin gamma1 causes apoptosis and prevents proliferation. J Neurosci, 2005. 25(18): p. 4463-72.  276.  Feltri, M.L. and L. Wrabetz, Laminins and their receptors in Schwann cells and hereditary neuropathies. J Peripher Nerv Syst, 2005. 10(2): p. 128-43.  277.  Fernandez-Valle, C., et al., Anti-beta 1 integrin antibody inhibits Schwann cell myelination. J Neurobiol, 1994. 25(10): p. 1207-26.  278.  Brown, N.H., S.L. Gregory, and M.D. Martin-Bermudo, Integrins as mediators of morphogenesis in Drosophila. Dev Biol, 2000. 223(1): p. 1-16.  279.  Schwabe, T., et al., GPCR signaling is required for blood-brain barrier formation in drosophila. Cell, 2005. 123(1): p. 133-44.  280.  Sun, B., P. Xu, and P.M. Salvaterra, Dynamic visualization of nervous system in live Drosophila. Proc Natl Acad Sci U S A, 1999. 96(18): p. 10438-43.  281.  Sepp, K.J. and V.J. Auld, Conversion of lacZ enhancer trap lines to GAL4 lines using targeted transposition in Drosophila melanogaster. Genetics, 1999. 151(3): p. 1093-101.  282.  Lee, T. and L. Luo, Mosaic analysis with a repressible cell marker for studies of gene function in neuronal morphogenesis. Neuron, 1999. 22(3): p. 451-61.  283.  Page-McCaw, A., et al., Drosophila matrix metalloproteinases are required for tissue remodeling, but not embryonic development. Dev Cell, 2003. 4(1): p. 95-106.  284.  Dietzl, G., et al., A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila. Nature, 2007. 448(7150): p. 151-6.  285.  McGuire, S.E., et al., Spatiotemporal rescue of memory dysfunction in Drosophila. Science, 2003. 302(5651): p. 1765-8.  286.  Bunch, T.A., et al., Characterization of mutant alleles of myospheroid, the gene encoding the beta subunit of the Drosophila PS integrins. Genetics, 1992. 132(2): p. 519-28.  287.  Kelso, R.J., et al., Flytrap, a database documenting a GFP protein-trap insertion screen in Drosophila melanogaster. Nucleic Acids Res, 2004. 32(Database issue): p. D418-20.  288.  Morin, X., et al., A protein trap strategy to detect GFP-tagged proteins expressed from their endogenous loci in Drosophila. Proc Natl Acad Sci U S A, 2001. 98(26): p. 15050-5.  289.  Alfonso, T.B. and B.W. Jones, gcm2 promotes glial cell differentiation and is required with glial cells missing for macrophage development in Drosophila. Dev Biol, 2002. 248(2): p. 369-83.  290.  Wada, A., et al., Specialized extraembryonic cells connect embryonic and extraembryonic epidermis in response to Dpp during dorsal closure in Drosophila. Dev Biol, 2007. 301(2): p. 340-9. 166  291.  Karpova, N., et al., Jupiter, a new Drosophila protein associated with microtubules. Cell Motil Cytoskeleton, 2006. 63(5): p. 301-12.  292.  Delon, I. and N.H. Brown, Integrins and the actin cytoskeleton. Curr Opin Cell Biol, 2007. 19(1): p. 43-50.  293.  Tavernarakis, N., et al., Heritable and inducible genetic interference by double-stranded RNA encoded by transgenes. Nat Genet, 2000. 24(2): p. 180-3.  294.  Xiong, W.C., et al., repo encodes a glial-specific homeo domain protein required in the Drosophila nervous system. Genes Dev, 1994. 8(8): p. 981-94.  295.  Bokel, C. and N.H. Brown, Integrins in development: moving on, responding to, and sticking to the extracellular matrix. Dev Cell, 2002. 3(3): p. 311-21.  296.  Narasimha, M. and N.H. Brown, Novel functions for integrins in epithelial morphogenesis. Curr Biol, 2004. 14(5): p. 381-5.  297.  Llano, E., et al., Structural and enzymatic characterization of Drosophila Dm2-MMP, a membrane-bound matrix metalloproteinase with tissue-specific expression. J Biol Chem, 2002. 277(26): p. 23321-9.  298.  Schmid, A., A. Chiba, and C.Q. Doe, Clonal analysis of Drosophila embryonic neuroblasts: neural cell types, axon projections and muscle targets. Development, 1999. 126(21): p. 4653-89.  299.  Corfas, G., et al., Mechanisms and roles of axon-Schwann cell interactions. J Neurosci, 2004. 24(42): p. 9250-60.  300.  Ruppert, M., et al., The L1 adhesion molecule is a cellular ligand for VLA-5. J Cell Biol, 1995. 131(6 Pt 2): p. 1881-91.  301.  Montgomery, A.M., et al., Human neural cell adhesion molecule L1 and rat homologue NILE are ligands for integrin alpha v beta 3. J Cell Biol, 1996. 132(3): p. 475-85.  302.  Itoh, K., et al., Disrupted Schwann cell-axon interactions in peripheral nerves of mice with altered L1-integrin interactions. Mol Cell Neurosci, 2005. 30(4): p. 624-9.  303.  Eroglu, C. and B.A. Barres, Regulation of synaptic connectivity by glia. Nature, 2010. 468(7321): p. 223-31.  304.  Clegg, D.O., et al., Integrins in the development, function and dysfunction of the nervous system. Front Biosci, 2003. 8: p. d723-50.  305.  Brakebusch, C. and R. Fassler, The integrin-actin connection, an eternal love affair. EMBO J, 2003. 22(10): p. 2324-33.  306.  Critchley, D.R., et al., Integrin-mediated cell adhesion: the cytoskeletal connection. Biochem Soc Symp, 1999. 65: p. 79-99.  307.  Critchley, D.R., Focal adhesions - the cytoskeletal connection. Curr Opin Cell Biol, 2000. 12(1): p. 133-9.  308.  Yu, W.M., et al., Laminin is required for Schwann cell morphogenesis. J Cell Sci, 2009. 122(Pt 7): p. 929-36.  309.  Benninger, Y., et al., Essential and distinct roles for cdc42 and rac1 in the regulation of Schwann cell biology during peripheral nervous system development. J Cell Biol, 2007. 177(6): p. 1051-61.  310.  Grove, M., et al., FAK is required for axonal sorting by Schwann cells. J Cell Biol, 2007. 176(3): p. 167  277-82. 311.  Li, M. and D.S. Sakaguchi, Expression patterns of focal adhesion associated proteins in the developing retina. Dev Dyn, 2002. 225(4): p. 544-53.  312.  Previtali, S.C., et al., Role of integrins in the peripheral nervous system. Prog Neurobiol, 2001. 64(1): p. 35-49.  313.  Murakami, S., et al., Focal adhesion kinase controls morphogenesis of the Drosophila optic stalk. Development, 2007. 134(8): p. 1539-48.  314.  Kandachar, V., T. Bai, and H.C. Chang, The clathrin-binding motif and the J-domain of Drosophila Auxilin are essential for facilitating Notch ligand endocytosis. BMC Dev Biol, 2008. 8: p. 50.  315.  Hay, B.A., T. Wolff, and G.M. Rubin, Expression of baculovirus P35 prevents cell death in Drosophila. Development, 1994. 120(8): p. 2121-9.  316.  Michelson, A.M., et al., Heartbroken is a specific downstream mediator of FGF receptor signalling in Drosophila. Development, 1998. 125(22): p. 4379-89.  317.  Rees, J.S., et al., In vivo analysis of proteomes and interactomes using Parallel Affinity Capture (iPAC) coupled to mass spectrometry. Mol Cell Proteomics, 2011. 10(6): p. M110 002386.  318.  Fujita, S.C., et al., Monoclonal antibodies against the Drosophila nervous system. Proc Natl Acad Sci U S A, 1982. 79(24): p. 7929-33.  319.  Goto, S. and S. Hayashi, Proximal to distal cell communication in the Drosophila leg provides a basis for an intercalary mechanism of limb patterning. Development, 1999. 126(15): p. 3407-13.  320.  Xie, X. and V.J. Auld, Integrins are necessary for the development and maintenance of the glial layers in the Drosophila peripheral nerve. Development, 2011. 138(17): p. 3813-22.  321.  Milner, R., et al., Distinct roles for astrocyte alphavbeta5 and alphavbeta8 integrins in adhesion and migration. J Cell Sci, 1999. 112 ( Pt 23): p. 4271-9.  322.  Etienne-Manneville, S. and A. Hall, Integrin-mediated activation of Cdc42 controls cell polarity in migrating astrocytes through PKCzeta. Cell, 2001. 106(4): p. 489-98.  323.  Berti, C., et al., Non-redundant function of dystroglycan and beta1 integrins in radial sorting of axons. Development, 2011. 138(18): p. 4025-37.  324.  Baron, W., H. Colognato, and C. ffrench-Constant, Integrin-growth factor interactions as regulators of oligodendroglial development and function. Glia, 2005. 49(4): p. 467-79.  325.  Afshari, F.T., et al., Schwann cell migration is integrin-dependent and inhibited by astrocyte-produced aggrecan. Glia, 2010. 58(7): p. 857-69.  326.  Eliceiri, B.P., Integrin and growth factor receptor crosstalk. Circ Res, 2001. 89(12): p. 1104-10.  327.  Rusnati, M., et al., alphavbeta3 integrin mediates the cell-adhesive capacity and biological activity of basic fibroblast growth factor (FGF-2) in cultured endothelial cells. Mol Biol Cell, 1997. 8(12): p. 2449-61.  328.  Mori, S., et al., Direct binding of integrin alphavbeta3 to FGF1 plays a role in FGF1 signaling. J Biol Chem, 2008. 283(26): p. 18066-75.  329.  Tsou, R. and F.F. Isik, Integrin activation is required for VEGF and FGF receptor protein presence on human microvascular endothelial cells. Mol Cell Biochem, 2001. 224(1-2): p. 81-9.  330.  Subramanian, A., et al., Thrombospondin-mediated adhesion is essential for the formation of the 168  myotendinous junction in Drosophila. Development, 2007. 134(7): p. 1269-78. 331.  Fraichard, S., et al., Tenectin is a novel alphaPS2betaPS integrin ligand required for wing morphogenesis and male genital looping in Drosophila. Dev Biol, 2010. 340(2): p. 504-17.  332.  Borchiellini, C., J. Coulon, and Y. Le Parco, The function of type IV collagen during Drosophila muscle development. Mech Dev, 1996. 58(1-2): p. 179-91.  333.  Mirouse, V., et al., Dystroglycan and perlecan provide a basal cue required for epithelial polarity during energetic stress. Dev Cell, 2009. 16(1): p. 83-92.  334.  Schneider, M., et al., Perlecan and Dystroglycan act at the basal side of the Drosophila follicular epithelium to maintain epithelial organization. Development, 2006. 133(19): p. 3805-15.  335.  Itoh, K., et al., Brain development in mice lacking L1-L1 homophilic adhesion. J Cell Biol, 2004. 165(1): p. 145-54.  336.  Potter, C.J., et al., The Q system: a repressible binary system for transgene expression, lineage tracing, and mosaic analysis. Cell, 2010. 141(3): p. 536-48.  337.  Lai, S.L. and T. Lee, Genetic mosaic with dual binary transcriptional systems in Drosophila. Nat Neurosci, 2006. 9(5): p. 703-9.  338.  Dickson, B.J., Molecular mechanisms of axon guidance. Science, 2002. 298(5600): p. 1959-64.  339.  Yiu, G. and Z. He, Glial inhibition of CNS axon regeneration. Nat Rev Neurosci, 2006. 7(8): p. 617-27.  340.  Raper, J.A., Semaphorins and their receptors in vertebrates and invertebrates. Curr Opin Neurobiol, 2000. 10(1): p. 88-94.  341.  Yu, L., et al., Plexin a-semaphorin-1a reverse signaling regulates photoreceptor axon guidance in Drosophila. J Neurosci, 2010. 30(36): p. 12151-6.  342.  Cafferty, P., et al., Semaphorin-1a functions as a guidance receptor in the Drosophila visual system. J Neurosci, 2006. 26(15): p. 3999-4003.  343.  Newsome, T.P., B. Asling, and B.J. Dickson, Analysis of Drosophila photoreceptor axon guidance in eye-specific mosaics. Development, 2000. 127(4): p. 851-60.  344.  Yu, H.H., A.S. Huang, and A.L. Kolodkin, Semaphorin-1a acts in concert with the cell adhesion molecules fasciclin II and connectin to regulate axon fasciculation in Drosophila. Genetics, 2000. 156(2): p. 723-31.  345.  Wong, J.T., S.T. Wong, and T.P. O'Connor, Ectopic semaphorin-1a functions as an attractive guidance cue for developing peripheral neurons. Nat Neurosci, 1999. 2(9): p. 798-803.  346.  Lattemann, M., et al., Semaphorin-1a controls receptor neuron-specific axonal convergence in the primary olfactory center of Drosophila. Neuron, 2007. 53(2): p. 169-84.  347.  Sweeney, L.B., et al., Temporal target restriction of olfactory receptor neurons by Semaphorin-1a/PlexinA-mediated axon-axon interactions. Neuron, 2007. 53(2): p. 185-200.  169  Appendix A  Integrins are required for glia proliferation in  the optic stalk A.1 Integrin RNAi in glia gives a thinner optic stalk phenotype I observed a range of glial phenotypes in the RNAi active optic stalks and eye disc. I generally summarized the glial phenotypes into two groups based on the changes to the optic stalk morphology: thinner and thicker.  When compared to controls at the  wandering L3 stage (Fig. A. 1A), the thinner optic stalk is characterized by a smaller diameter (Fig. A. 1B-C). The severity and penetrance of the thinner and thicker optic stalk phenotypes varied in experiments in which different RNAi lines were used or other experimental conditions that altered the efficiency of the RNAi (UAS-Dicer2 and temperature)(Table A. 1). Three UAS-βPS-RNAi lines were used in our experiment: KK and GD lines from the VDRC, and a R1 line from the NIG (see materials and methods in Chapter 3).  Under  standard experimental conditions, I observed the thinner optic stalk phenotype predominantly using the βPS-RNAi(KK) line (n=40) and βPS-RNAi(GD) (n=48) while the majority of the progeny from βPS-RNAi(R1) line had the thicker optic stalk phenotype (56/59).  Even though both gave thinner optic stalk, the βPS-RNAi(KK) and  (GD) lines did show some differences.  For example, the GD line consistently had  slightly more glia in the optic stalk than the KK line (Fig. A. 2A) and the optic stalk 170  shapes were not always the same between these two RNAi lines (Fig. A. 1B-C). The phenotypic differences between the different RNAi lines are likely due to different RNAi efficiency of knocking down βPS integrin.  I tested this by reducing the efficiency of the  βPS-RNAi(GD) line by eliminating UAS-Dicer2 from the cross or lowering the temperature from 25 °C to 18 °C. Under these conditions the GD line yielded more of the thicker optic stalk (Table A. 1), which were similar to the repo>βPS-RNAi(R1) optic stalks (data not shown).  171  Figure A. 1: Thinner optic stalk phenotype A repo-GAL4 was used to express βPS-RNAi lines in all glia with a CD8::GFP (green) to label glial membrane.  Dicer2 was co-expressed to facilitate RNAi.  Glial  nuclei and photoreceptor cells were labeled by Repo (red) and HRP (blue) immunolabeling, respectively. Note that the HRP channel is not necessarily at the same focal plane as other color channels.  When comparing with the control (A),  repo>βPS-RNAi (KK or GD) larvae had thinner optic stalks (B&C, arrows) in which far fewer glia were observed with Repo immunolabeling.  All images were taken using a  40X air lens with an Axioskop 2 system (Zeiss). The unequal eye disc sizes are due to different developmental stages and genotypes. .  172  Optic stalk phenotype Genotypes repo-GAL4 repo>βPS-RNAi(KK) repo>βPS-RNAi(GD) repo>βPS-RNAi(GD) (at 18 °C) *repo-GAL4 *repo>βPS-RNAi(GD)  Axonal phenotype classes  Normal  Thicker  Thinner  Total  Normal  Mis-targeting  Stalling in OS  Stalling in ED  Total  29 0 0  0 0 0  0 40 48  29 40 48  40 0 0  0 24 27  0 0 7  0 4 0  40 28 34  8  46  0  54  46  8  0  0  54  22 3  0 41  0 0  22 44  32 14  0 13  0 5  0 10  32 42  Table A. 1. Summary of glial and neuronal phenotypes. All the genotypes that were examined in the text for the optic stalk morphologic changes, photoreceptor axon outgrowth and targeting phenotypes are listed. All RNAi and control crosses were done with co-expression of UAS-CD8-GFP and at 25°C unless specified. UAS-Dicer2 was co-expressed in the crosses that were not marked with (*).  173  Abbreviations: ED - eye disc; OS -optic stalk.  A.2 Thinner optic stalk is due to decreased glial proliferation Far fewer glia were observed within the thinner optic stalks compared to controls (Fig. A. 1B-C). To confirm this observation, glial number was counted in early and wandering L3 optic stalks plus the eye disc (Fig. A. 2A).  The glial number in  repo>βPS-RNAi(KK) and repo>βPS-RNAi(GD) larvae decreased to less than 20% of that in the controls at both stages.  Normally in early L3 larvae, glia are still found  exclusively within the optic stalk and have not entered the eye disc.  At this stage, on  average, 40±12 glia were found in control optic stalks (n=36), while 7±2 glia were seen in the repo>βPS-RNAi(GD) (n=42) optic stalks and 5±1 glia in the repo>βPS-RNAi(KK) (n=36) optic stalks (Fig. A. 2A).  After photoreceptor cells begin to form at the middle  3rd instar stage, glia migrate into the eye disc and the glial number in the eye disc is positively related to the age of the eye disc [126]. To control for variability in eye disc age, wandering L3 eye discs with 6-10 rows of ommatidia were scored for glial number. At this stage, each control optic stalk plus eye disc had 188±20 (n=29) glia while the repo>βPS-RNAi(GD) and repo>βPS-RNAi(KK) larvae had only 14±6 (n=32) and 9±5 (n=30) glia in each optic stalk plus eye disc, respectively (Fig. A. 2A).  Our data  suggested that the thinner phenotype was caused by a dramatic reduction in glia number in the optic stalk and eye disc. proliferation and cell death.  The loss of glia could result from changes to cell  However, TUNEL staining in repo>βPS-RNAi(GD) and 174  repo>βPS-RNAi(KK) did not show increased glial apoptosis compared to the controls in early L3 optic stalks (data not shown).  Co-expression of an apoptosis inhibitor,  baculovirus p35 [315] did not significantly increase glia in these RNAi larvae (data not shown), suggesting cell death is not the reason for the loss of glia. To address whether there was a reduction in glial proliferation, I used a phosphorylated Histone-3 (PH3) antibody to label mitotic cells and scored the dividing glial cells in early L3 optic stalks.  At this stage all the glia are within the optic stalk and  have not entered the eye disc (Fig. A. 2D).  Dividing cells positive for PH3  immunolabeling were seen in both the optic stalk and eye disc.  When specifically  assayed for Repo immunolabeling, actively dividing glial nuclei were only seen in the optic stalk at this stage (Fig. A. 2B).  In the control (n=36), there were 1.3±1.0 PH3  positive glia in the optic stalks and dividing glia were seen in 81% of control optic stalks (Fig. A. 2C).  While in the repo>βPS-RNAi(GD) (n=44), the optic stalks had 0.2±0.5  PH3 positive nuclei on average and only 16% of the optic stalks had dividing glia (Fig. A. 2C).  Moreover, there was no PH3 staining detected in any of the repo>βPS-RNAi(KK)  optic stalks that I scored (n=36).  The PH3 labeled dividing glial number was consistent  with the overall number of glia in the controls compared to the repo>βPS-RNAi(KK or GD) optic stalks.  This suggests that the thinner optic stalk phenotype is due to a lack of  glial proliferation and βPS integrin is required for glial proliferation in the optic stalk. 175  Although having a thicker optic stalk, the repo>βPS-RNAi(R1) also had slightly fewer glia in the eye disc plus optic stalk when compared to the control (Fig. A. 2A). The difference between RNAi lines was probably due to the different RNAi efficiency as discussed above.  However, off-target effects are always a concern.  To confirm the  specificity of integrins on glial proliferation, I generated somatic MARCM clones specifically in glia using the null allele of mys1 as I had done previously using repo-flipase to generate glial specific clones [320].  In the wL3 larvae, control eye disc  and optic stalk (n=20) had 19±18 CD8-GFP labeled glia while the βPS (mys) mutant clones (n=50) were significantly smaller consisting of 6±4 glia. idea that βPS integrin has a role in glial proliferation.  176  The results support the  177  Figure A. 2: Decreased glia proliferation and migration in repo>βPS-RNAi optic stalks. Comparisons of glia number (A&E), glia proliferation (B&C) between the control and βPS LOF (RNAi or MARCM) in the optic stalk and eye disc.  Kruskal-Wallis test  with Dunn’s Multiple Comparison post-test was used to compare either RNAi genotype to the control (A&C).  Unpaired T-test was used to compare MARCM genotypes (E).  All comparisons had P<0.001 as indicated by three stars.  Error bars stand for standard  deviation. Numbers show the mean value for each genotype. A) Total glial number was quantified in the eye disc (ED) plus optic stalk (OS) by Repo immunolabeling from the control (repo-GAL4: red) and two repo>βPS-RNAi genotypes (GD: blue and KK: green) in two developmental stages (eL3-early 3rd instar and wL3-wandering 3rd instar with 6-10 rows of ommatidia).  For simplification,  genotypes were labeled with the names of the RNAi lines that repo-GAL4 crossed with. B) Phosphorylated-Histone 3 (PH3, red) immuolabeling was used to identify dividing cells in eL3 optic stalks. Glial nuclei were labeled with a Repo antibody (green) and all nuclei were shown by DAPI staining (blue). Dividing cells were found in both the eye disc (arrowheads) and optic stalk (arrow). Notice that glial number was greatly reduced and no dividing glia were found in the βPS-RNAi optic stalk (right). Both images are projections of Z-stacks.  Scale bars are 15 μm.  C) Quantification of dividing glial number in the eL3 optic stalk. The left panel compares the average number of PH3 labeled dividing glial per optic stalk and the right panel compares the percentage of the optic stalk that had dividing glia. D) A cartoon to show the architecture of an early 3 rd instar optic stalk and eye disc. The glial proliferation (the red cell) is restricted in the optic stalk since no glia migrates into the eye disc at this stage. 178  E) MARCM was used to generate wild-type (WT) or homozygous βPS glial clones. Representative images (left) show glial clones (CD8::GFP, green) and photoreceptor cells (HRP labeling, red) in the wL3 eye discs and optic stalks. quantified and compared in the right panel.  179  Glial clone size was  A.3 Discussion The variety of glial and neuronal phenotypes in the repo>βPS-RNAi and repo>talin-RNAi allowed us to characterize the role of integrins in the different stages of glial development in the optic stalk.  I found that βPS integrin was required at early  larval stages for PG proliferation in the optic stalk.  Strong knock-down of βPS integrin  resulted in reduced glial proliferation generating a significantly smaller glial population in the optic stalk and eye disc. The variety of glial phenotypes that I observed was possibly dependent on RNAi efficiencies with different UAS-βPS-RNAi lines or at different experimental conditions. RNAi efficiency could affect protein expression in a temporal or quantitative manner, which I was unable to differentiate between. Thus our data indicated that βPS integrin might play early roles of promoting PG proliferation, organization and CG barrier formation all through 2nd and 3rd instar larval stages, and late roles of regulating PG and photoreceptor axon migration in the 3rd instar.  Alternatively these glial activities could  require different levels of integrin complexes.  A low level of βPS integrin might be  sufficient to promote glial proliferation.  This conclusion is support by the reduced  levels of proliferation observed in the somatic clones of the null allele of PS.  This may  explain why no dramatic glial loss was seen in the weak βPS-RNAi or talin-RNAi larvae. While a high level of βPS integrin might be necessary for glial movement, which was 180  deficient across all RNAi lines.  Meanwhile, down-regulation of Talin in glia only gave  a subset of βPS integrin phenotypes, suggesting that Talin, although one of the major components in integrin adhesion complexes, might not be involved in all integrin signaling pathways in glia.  However, it is possible that off-target effects contributed to  the extreme lack of glia observed in the two strong βPS-RNAi lines because the thinner optic stalk phenotype was not seen in UAS-Dicer2/mys1;; repo-Gal4/UAS-βPS-RNAi (R1).  181  Appendix B  Semaphorin 1a expression in glia of the eye disc  and optic stalk B.1 Correlation between axonal stalling and abnormal glia deposition and differentiation Axonal migration is regulated by a range of extracellular environmental factors including, extracellular matrix proteins, adhesion molecules and secreted guidance cues. Axonal growth and regeneration can be inhibited by certain repulsive and inhibitory cues under physiological and pathological conditions [338, 339]. To investigate if RNAi treated glia were expressing inhibitors of axon guidance, we analyzed a range of known axon guidance proteins.  Semaphorin-1a (Sema1a) drew particular attention for us.  Sema1a belongs to the Semaphorin family, which is well known for being both a repulsive and attractive axonal guidance cue [340].  In the Drosophila visual system,  Sema1a has been suggested to act as a Plexin-1 receptor and regulate photoreceptor axonal fasciculation and targeting [341, 342].  However, we found that Sema1a was not  only expressed by neurons (Fig. B. 1A), but also expressed by glial cells in the eye disc and optic stalk. Sema1a endogenously tagged with GFP was present in WG cell bodies and their membrane processes (Fig. B. 1A&B).  Conversely, only low levels of  Sema1a::GFP were seen in the surrounding PG (Fig. B. 1A). 182  The Sema1a::GFP  localization is consistent with a Sema1a antibody immunolabeling (data not shown). The role of Sema1a in the WG is not known, and since photoreceptor axons from the same ommatidia are bundled together and ensheathed by WG membranes [130, 343], it is possible that Sema1a might act as a repulsive cue to help separate the axon fascicules from different ommatidia. When photoreceptor axons were stalled in the optic stalk, we observed that clusters of glia formed a cap-like structure posterior to the axonal stalling region instead of forming a tubular structure around the axons (Fig. B. 1D-E).  Adjacent to the axonal  terminals in the glia cap, a variable number of glial cells (from one to several) were found to express relatively high amount of Sema1a::GFP (Fig. B. 1D-F), which made them distinct from other PG (Fig. B. 1D-E).  These Sema1a positive glia were of interest  because the occurrence of these glia was highly correlated with the axonal stalling phenotype.  These glia were found in most of the repo::βPS-RNAi(R1) stalling optic  stalks (n=7/8), but never seen in the control optic stalks (n=8) or the repo::βPS-RNAi(R1) non-stalling optic stalks (n=8) (Fig. B. 1C).  The next questions were whether and how  these glia blocked axonal migration especially given that Sema1a is well known for its repulsive properties [344-347].  Rescue experiments were conducted by co-expressing  Sema1a-RNAi with βPS-RNAi in the glia.  However, we did not observe any consistent  rescue with three separate Sema1a-RNAi lines from multiple sources (data not shown). 183  This suggested that Sema1a might not be the only factor and other proteins might also be present to inhibit axonal growth. Another question that we addressed was the origin of the Sema1a positive glia. Since WG express high amount of Sema1a, could the cap-like region be formed from WG? However, the cap-like glia were different from normal WG in several ways. First, they were usually found prior to the stalled axonal terminals while WG always attach to and follow axons.  Second, these glia did not extend long membrane processes which  are a hallmark for WG.  Third, they were not positive for the Gli-lacZ marker (Fig. 6D-E,  asterisks) that reflects the expression of the Gliotactin gene, which is only detected in fully differentiated WG nuclei (Fig. 6B-E, arrowheads).  The differences suggested that  these Sema1a positive glia were unlikely to arise from the WG.  184  185  Figure B.1 Photoreceptor axon stalling phenotype is correlated with aberrant Sema1a and Gli-lacZ expression in the optic stalk. Semaphorin-1a::GFP (Sema1a::GFP, green in A-F) and Gli-lacZ (blue in C-F) are markers for differentiated wrapping glia (WG) in 0.2μm sections of control (A&C), repo>sema1a-RNAi (B) and repo>βPS-RNAi (D-F) wandering 3rd instar optic stalks. All glia membranes were labeled by repo>CD8::RFP.  Regions in white boxes were  digitally expanded and shown in right small panels. A) In a wild type larva, WG were distinct for their long membrane processes that paralleled with 24B10 immunolabeling photoreceptor axons (blue) in the central region of the optic stalk.  Compared to surrounding perineurial glia (PG) and axons, relatively  high level of Sema1a::GFP was found in WG cell bodies and membrane processes (arrowheads).  Note that strong Sema1a::GFP expression was also detected in the eye  disc (arrows). B) In a repo>sema1a-RNAi larva, Sema1a::GFP was greatly decreased in wrapping glia in the optic stalk (arrowheads) but seemed normal in photoreceptor cells in the eye disc (arrows). C) In all control OS (n=8), WG were simultaneously labeled by membranous Sema1a::GFP and nuclear Gli-lacZ (arrowheads). D) In some repo>βPS-RNAi(R1) OS (n=8), photoreceptor axons succeeded to penetrate the OS.  These optic stalks had morphologic changes, such as: multiple layers  of PG (triangles).  However, Sema1a::GFP and Gli-lacZ expression seemed to be normal  in these bigger optic stalks. E-F) Photoreceptor axons failed to exit the ED (n=4) or pass through the OS (n=8) in some repo>βPS-RNAi(R1).  In most of the stalling OS (n=7/8), glial cells were observed  in the posterior tip of stalling regions and expressed elevated levels of Sema1a::GFP 186  (asterisks), which made them distinct from other surrounding PG (triangles). In about half of these OS (n=3/7), these tip glia also expressed various amount of Gli-lacZ (asterisks in E).  Note that high level of Sema1a::GFP and Gli-lacZ were also expressed by WG that  were posterior to the stalling regions (arrowheads). G) The Sema1a fluorescent intensity in D and E was depicted using a ‘fire’ lookup table.  Arrowheads and asterisks point to the glia with high level of Sema1a::GFP in the  center of the stalling OS. Scale bars are 10 μm in left big panels and 5 μm in right three small panels.  187  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items